A Phase III Clinical Trial of a Haemophilus Influenzae Type b (Hib) Conjugate Vaccine: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Phase III Clinical Trial of a Haemophilus Influenzae Type b (Hib) Conjugate Vaccine (NCT02692859), a randomized phase 3 study. The primary outcome is Number of Participants With Solicited Adverse Reactions.

2. Clinical Question

The posted primary counts are 507 participants in the Chengdu Olymvax arm and 480 in the Walvax arm, with 996 participants listed for each arm. These counts describe participants with solicited adverse reactions, not the number of individual reactions. A participant may have had more than one reaction, but the posted primary measure counts participants rather than reaction episodes.

The record also names secondary outcomes concerning unsolicited adverse reactions, serious adverse events, and anti-PRP concentrations and responses. Those outcomes address aspects of safety and immunogenicity beyond the primary solicited-reaction count. The results for the named secondary outcomes are not reported in the registry record as numerical comparisons. The available primary outcome therefore answers a specific short-term safety question; it does not, by itself, describe the full set of outcomes named for the study.

3. Trial Design

This is a Phase III, randomized, positive-controlled, parallel-group trial. The registry lists triple masking. It identifies the two arms by the vaccine associated with each manufacturer, and each arm description names a Hib conjugate vaccine. The record does not specify which roles were masked or how masking was implemented. It also does not give details of the randomization procedure, allocation concealment, vaccine schedule, dose, or follow-up procedures beyond the primary outcome timeframe.

The registry reports enrollment of 1992. The posted outcome data list 996 participants in each arm. The trial status is completed, with a study start of 2013-09 and primary completion of 2014-08. The sponsor is Jiangsu Province Centers for Disease Control and Prevention. The study summary describes healthy infants and children in China as the intended population and frames the study as an evaluation of safety and immunogenicity.

Random allocation is a design approach intended, in general, to reduce systematic differences between groups at the start of a study. Parallel-group designs assign participants to an arm and compare outcomes across the arms. Masking is intended, in general, to limit the influence of knowledge about assignment on aspects of study conduct or outcome assessment. These are general descriptions of design concepts; the registry record does not provide enough procedural detail to assess how these features were carried out in this trial.

The official title calls the trial non-inferiority, while the posted primary outcome is the number of participants with solicited adverse reactions. The registry record does not report the planned non-inferiority margin, the formal hypothesis or decision rule, or a sponsor-reported non-inferiority result. The posted comparison can be described as a comparison of the observed primary counts, but that information alone does not establish whether the trial met a prespecified non-inferiority criterion.

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.

Overall Study

Analysis population for the primary outcome, as defined in the registry: A total of participants in aged 3-5 months,6-11months and 1-5years

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 Number of Participants With Solicited Adverse Reactions, measured over 0-7 days after each dose.

The primary endpoint is the number of participants with solicited adverse reactions, described as a count of participants with such reactions during 0-7 days after each dose. The outcome description refers to participants in age groups 3-5 months, 6-11months and 1-5years. The posted arm-level values are 507 of 996 for the Chengdu Olymvax vaccine and 480 of 996 for the Walvax vaccine.

This endpoint records whether a participant had at least one solicited adverse reaction in the stated observation window. It is not a count of every reaction, and it does not show the number, severity, duration, or timing of individual reactions. Those details are not reported in the registry record. The outcome is a binary participant-level event summarized as a count in each arm.

Solicited adverse reactions are a safety outcome that can be compared between arms over a defined period after dosing. In general, a prespecified observation window helps make the timing of outcome assessment clear. The registry record does not define which reactions were solicited, how they were elicited, or how they were graded. It also does not provide separate primary results by the age ranges named in the outcome description.

The endpoint is relevant to the trial’s stated safety evaluation, but it is not a complete account of safety. The record separately names unsolicited adverse reactions and serious adverse events as secondary outcomes. Results for those outcomes, along with the immunogenicity outcomes, are not reported in the registry record as comparative numerical results.

7. Results Posted in the Registry

• Vaccine (Walvax Biotechnology Co., LTD.): 48.2% (480/996) versus 50.9% (507/996) on control; risk difference -2.7 percentage points (95% CI -7.1 to 1.7), risk ratio 0.95 (0.87–1.03); unadjusted two-proportion z-test p = 0.226.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is a participant-level event count. Each participant is classified according to whether a solicited adverse reaction was recorded in the stated window. The registry reports counts and arm sizes, while the comparison includes an unadjusted risk difference and a p-value. The statistical explanation below describes these concepts generally and distinguishes the posted comparison from the trial’s unreported formal analysis plan.

What does a participant count represent?

A participant count records how many people met the outcome definition at least once. It does not count repeated events separately. In this trial, the definition is the presence of a solicited adverse reaction during 0-7 days after each dose. The registry does not identify the list of solicited reactions or say how multiple doses or multiple reactions within a participant were handled in the final summary. The count should therefore be interpreted according to the stated participant-level outcome, rather than as a detailed event history.

In general, binary outcomes can be summarized by the number of participants with an event and the number assessed. A risk difference compares the event proportions between groups by subtracting one group’s proportion from the other’s. Its direction depends on the order of groups used in the comparison. The posted comparison identifies the Walvax arm against the Chengdu Olymvax arm; a negative difference therefore points toward a lower observed event proportion in the Walvax arm under that ordering. This is a statistical description, not a claim about the clinical importance of the difference.

How should the p-value be read?

The posted unadjusted p-value is 0.226. In general, a p-value is calculated under a null hypothesis and a statistical model. It describes how unusual data at least as discrepant as the observed data would be if that null model and its assumptions held. 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. A p-value alone also does not show that groups are equivalent.

The registry does not identify the test underlying the posted p-value or provide a detailed analysis specification. The comparison is labelled unadjusted, so it should not be presented as adjusted for age, baseline characteristics, or other variables. Whether such adjustments were planned or performed is not reported in the registry record. The recalculated summary comparison is unadjusted; the registry does not provide an additional sponsor analysis of this primary outcome in the registry record.

What does a confidence interval contribute?

A confidence interval gives a range of values compatible with the data and the method used to calculate it. In general, an interval communicates uncertainty around an estimate more directly than a point estimate alone. For the recalculated risk difference, the interval spans zero. Zero represents no difference on the risk-difference scale, so the interval includes values on both sides of no difference. This does not prove that there is no difference; it indicates that the interval does not isolate one direction.

Confidence intervals rely on statistical assumptions, and they describe uncertainty in the estimate rather than covering every possible source of error. They do not account automatically for bias from missing outcomes, measurement procedures, or departures from the analysis plan. The registry record does not report enough detail to evaluate these matters for this trial.

What is not available about the formal design analysis?

The title identifies a non-inferiority trial. In general, a non-inferiority analysis asks whether the difference is sufficiently limited relative to a prespecified margin. The margin, analysis population, confidence-bound decision rule, and sponsor-reported non-inferiority result are not reported in the registry record. The unadjusted risk-difference calculation is not evidence that the trial’s formal non-inferiority criterion was met.

The record lists zero posted analyses in the analysis count. That does not mean no statistical work was conducted; it means formal analysis details are not represented in the registry results information available for this account. The randomization method, handling of missing observations, covariate adjustment, subgroup methods, multiplicity procedure, and sensitivity analyses are not reported in the registry record. These omissions limit the ability to reproduce or fully assess the intended statistical approach.

The measure is a count outcome, but the event status of each participant makes the posted comparison a binary-outcome comparison. It is not an analysis of the number of reactions per participant or a time-to-event analysis. General statistical concepts can clarify the types of quantities involved, but they cannot replace the missing protocol and analysis details. Conclusions should therefore be confined to the arm-level participant counts and the unadjusted comparison that are posted.

10. Confidence Intervals and Precision

The posted primary results give the event count in each arm and the number listed for that arm: 507 of 996 in the Chengdu Olymvax arm and 480 of 996 in the Walvax arm. The comparison includes a recalculated risk-difference confidence interval. That interval spans zero, so it does not rule out either direction of the difference using this unadjusted calculation. Its breadth reflects uncertainty around the estimated difference, rather than a direct measure of clinical importance.

In general, a confidence interval is a range generated by a statistical procedure that, under its assumptions, expresses uncertainty about an underlying quantity. A narrower interval indicates greater precision for that quantity under that method; a wider interval indicates less precision. An interval should be interpreted together with its scale, the analysis population, and the method used to construct it. The posted interval is for an unadjusted risk difference, not for a non-inferiority margin.

The posted p-value is 0.226. It does not substitute for an interval: it does not show the range of plausible differences or establish equivalence. Conversely, an interval that spans zero does not establish that the groups have exactly the same event frequency. The registry record does not report a prespecified clinical threshold for a meaningful difference, so the clinical importance of possible values within the interval cannot be judged from a stated threshold.

Precision is also limited by what the registry reports. The primary outcome is not broken out by the age ranges named in its description, and the registry does not report missing-outcome counts, analysis populations, or a detailed method for the interval. The posted result therefore gives a summary comparison, not a complete account of uncertainty arising from every design or data-quality issue. The formal non-inferiority margin and sponsor analysis are not reported in the registry record, preventing assessment of precision against that specific criterion.

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

The 80% power column gives the illustrative number of participants per arm for a design with an 80% chance of detecting the chosen effect under the stated assumptions, if that effect and the model assumptions were correct. The 90% power column gives the corresponding planning number for a design with a 90% chance under those same assumptions. These are design calculations, not results from the completed trial. They are not observed power.

The rows describe planning for 100% of the observed effect, 75% of the observed effect, and 50% of the observed effect. As the effect being targeted becomes smaller, more participants are required under the stated assumptions. The values are 5,339 and 7,147 participants per arm for planning at 100% of the observed effect; 9,493 and 12,708 for planning at 75%; and 21,360 and 28,595 for planning at 50%, for the 80% and 90% power columns respectively.

The plotted planning curve represents this general relationship between target effect and required sample size. It should be read as a conditional planning illustration, not as evidence that the observed study had a particular probability of finding a result. Because the calculation uses the observed effect as its reference, it depends on an effect estimate that is itself uncertain. The registry record does not report whether these illustrative assumptions were used to plan the trial.

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 = 0.509; treatment proportion p₁ = 0.482; average p̄ = 0.495.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.7071 = 1.3858.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.7068 = 0.5949.
4. n = (1.3858 + 0.5949)² / (-0.0271)² = 5,338.7, rounded up to 5,339 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 <- 0.509036144578313                          # control proportion
p1 <- 0.481927710843373                          # 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(2692859)
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 = 5,339 (80% power), 7,147 (90% power)
 75% of observed effect: n per arm = 9,493 (80% power), 12,708 (90% power)
 50% of observed effect: n per arm = 21,360 (80% power), 28,595 (90% power)
power.prop.test: 5338.7 
Simulated power at n = 5,339 per arm: 0.796 (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, number of participants with solicited adverse reactions, and names secondary outcomes for unsolicited adverse reactions, serious adverse events, anti-PRP concentration, geometric mean concentrations, and geometric mean fold increase. The record reports seven outcome measures, while the named outcome descriptions do not provide a full statistical analysis plan. The multiplicity strategy, including whether any adjustment was intended for multiple outcomes or comparisons, is not reported in the registry record.

Multiplicity matters in general because examining several outcomes or comparisons can increase the chance of at least one apparently notable result when each is interpreted separately. That general issue does not establish that the trial’s findings are false or invalid. It means that a set of results should be interpreted in light of prespecified outcome priorities and any planned adjustment. The registry does not state whether secondary outcomes were treated as confirmatory or descriptive, or how their results were intended to be interpreted.

The analysis count is listed as zero, and no detailed sponsor analysis is posted in the registry record for this account. The registry record does not report the analysis populations, whether an intention-to-treat, per-protocol, or other population was used, or how participants were handled if they did not complete follow-up. Missing-data counts, imputation methods, and sensitivity analyses are not reported. The number of randomized participants and the outcome denominators do not by themselves establish whether every randomized participant contributed an observed outcome or how missing results were handled.

The primary outcome description refers to age ranges, but the registry does not provide age-specific outcome counts or a subgroup analysis plan. It also does not report covariate adjustment or a plan for handling repeated observations across doses. These details are important for reproducing the analysis and understanding whether the summary reflects the intended estimand. Without them, interpretation should remain with the posted arm-level counts and the unadjusted comparison, without assuming an unreported method.

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.

Most frequent serious adverse event terms

Serious events with the highest arm-level rate: bronchopneumonia (1.2%), upper respiratory tract infection (0.1%), hand-foot-and-mouth disease (0.1%), pneumonia (0.1%), viral enteritis (0.1%).

The serious adverse event data list 10 in the Chengdu Olymvax arm and 15 in the Walvax arm, with 996 participants at risk in each arm. These are posted counts, not an adjusted comparison. The registry does not report a comparative estimate or confidence interval for serious adverse events, so the counts should be treated descriptively rather than as a statistical test of a difference.

The record lists bronchopneumonia at 1.2%, and upper respiratory tract infection, hand-foot-and-mouth disease, pneumonia, and viral enteritis at 0.1% each among the top serious-event terms. The available record does not establish from these figures alone how terms map to each arm, whether participants could contribute multiple terms, or how the percentages were calculated. Serious-event terms should not be interpreted as evidence that the vaccine caused the events.

The registry also lists a serious-event follow-up period as the whole study period among its secondary outcomes. Death counts are not reported in the registry record. Further detail on event timing, severity, relatedness assessment, and arm-specific term counts is not reported in the registry record. The serious adverse event counts and the primary solicited-reaction counts describe different outcomes and should not be combined.

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

The posted primary outcome shows that 507 participants in the Chengdu Olymvax arm and 480 in the Walvax arm were recorded as having solicited adverse reactions during 0-7 days after each dose. The registry lists 996 participants in each arm. The unadjusted comparison reports p = 0.226, and the recalculated risk-difference interval spans zero. Together, these results describe the observed counts and the uncertainty of an unadjusted comparison; they do not establish that the arms are identical.

The trial’s title identifies a non-inferiority design, but a non-inferiority conclusion requires a prespecified margin and an analysis interpreted against that margin. Neither the margin nor the sponsor’s formal non-inferiority analysis is reported in the registry record. The posted results therefore cannot establish whether the trial met its intended non-inferiority criterion. A non-significant p-value is not itself evidence of non-inferiority or equivalence.

The primary result is specific to solicited reactions and to the stated post-dose window. It does not characterize all possible safety outcomes, longer-term effects, or immunogenicity. Serious adverse event counts are posted separately, but the registry does not provide a formal comparative analysis for those counts in the available results. Numerical comparative results for the named secondary immunogenicity and unsolicited-reaction outcomes are not reported in the registry record.

In general, clinical interpretation also depends on the event definition, data completeness, study conduct, and the importance assigned to different outcomes. The registry record does not provide enough information about solicited-event definitions, missing data, analysis populations, or the detailed analysis plan to evaluate those aspects fully. The results can be described as the posted observations and an unadjusted statistical comparison, without extending them to claims about treatment benefit, causation, or outcomes that are not reported.

16. Why This Trial Matters Statistically

This trial illustrates why a trial title and a posted result need to be read together with the statistical details. The title calls the study non-inferiority, but the registry’s available numerical primary result is a participant count for solicited adverse reactions. Without the non-inferiority margin and formal decision rule, a reader cannot determine whether the observed comparison satisfies the design’s stated objective. A p-value alone cannot supply that missing criterion.

It also illustrates the difference between an event count and an event rate or event history. The endpoint counts participants with solicited adverse reactions in a defined post-dose window. It does not report how many separate reactions occurred, their severity, or their individual timing. Keeping the unit of analysis clear prevents the count from being overinterpreted as a complete summary of safety.

Randomized comparisons can support structured comparisons between assigned groups, but interpretation still depends on outcome definitions, follow-up, missing data, and analysis choices. Here, the registry reports arm counts and an unadjusted comparison, while several important details of the analysis plan are not reported. The distinction between posted numbers and unreported methods is central to a reproducible statistical reading.

Finally, the trial names safety and immunogenicity outcomes, but numerical results are available for only a limited part of that outcome set in the record described. Statistical interpretation must respect that boundary. Recognizing what is recorded, what is calculable from the posted summaries, and what is absent helps prevent an unadjusted result from being mistaken for a complete trial conclusion.

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 does not report the non-inferiority margin, formal decision rule, or sponsor-reported non-inferiority analysis. This is a major limitation for interpreting a trial whose title specifies non-inferiority. The posted unadjusted comparison does not resolve whether the study met that objective.

The primary outcome description does not define the solicited reactions, assessment method, severity grading, or handling of events across doses. Although the endpoint timeframe is stated as 0-7 days after each dose, dose schedule and follow-up details are not reported. Age ranges appear in the outcome description, but age-specific results are not reported in the registry record.

The record does not report the randomization method, allocation concealment, masking roles or procedures, analysis populations, missing-outcome counts, missing-data methods, or sensitivity analyses. It also does not report whether covariates were adjusted for, how repeated dose-related observations were handled, or how multiplicity was addressed. These absences limit assessment of the risk of bias and prevent full reproduction of the statistical analysis.

Numerical comparative results for most named secondary outcomes are not reported in the registry record. Serious adverse event counts are available, but a comparative analysis and detailed arm-specific term breakdown are not reported. Death counts are not reported in the registry record. The posted sample-size planning illustration is based on the observed effect and stated generic assumptions; the trial’s original design assumptions and sample-size justification are not reported.

These limitations do not change the posted counts, but they restrict the conclusions that can be drawn from them. The record supports a narrow description of the primary outcome and an unadjusted comparison, not a full assessment of the intended non-inferiority analysis or every safety and immunogenicity outcome.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: A Phase III Clinical Trial of a Haemophilus Influenzae Type b (Hib) Conjugate Vaccine — NCT02692859.
• Registry record last updated 2017-09-20; status last verified 2017-07.
• Results first posted 2017-08-21.
• 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 registry describes a completed, randomized, triple-masked, parallel-group Phase III trial of Hib conjugate vaccines in healthy infants and children in China. The sponsor is Jiangsu Province Centers for Disease Control and Prevention. The listed enrollment is 1992, with the study start in 2013-09 and primary completion in 2014-08.

The primary outcome is the number of participants with solicited adverse reactions during 0-7 days after each dose. The posted counts are 507 of 996 in the Chengdu Olymvax arm and 480 of 996 in the Walvax arm. The unadjusted comparison reports p = 0.226. The registry record does not report the non-inferiority margin, formal sponsor analysis, or detailed analysis plan.

Serious adverse event counts and several secondary outcome names are also recorded, but full comparative results for the secondary outcomes are not reported in the registry record. The available findings should be interpreted as posted descriptive counts and an unadjusted comparison, with the missing design and analysis details kept in view.