Pneumococcal Vaccine Booster Study in Healthy Children 12-18 Mths Old Previously Primed With the Same Vaccines: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Pneumococcal Vaccine Booster Study in Healthy Children 12-18 Mths Old Previously Primed With the Same Vaccines (NCT00547248), a randomized phase 3 study. The primary outcome is Number of Subjects Reporting Rectal Temperature Greater Than (>) the Cut-off.

2. Clinical Question

3. Trial Design

The registry describes a Phase 3, randomized, double-masked, parallel study with enrollment of 756. Its stated purpose was to assess safety and antibody response after a booster vaccination in children aged 12 to 18 months who had previously received the same vaccines in a primary course. The study was completed. The registry record lists GlaxoSmithKline as sponsor.

The booster phase involved four original groups. Children in the Philippines had been primed at 6-10-14 weeks of age and received either Synflorix or Prevenar with Tritanrix-HepB, Hiberix, and Polio Sabin. Children in Poland had been primed at 2-4-6 months of age and received either pneumococcal vaccine with Tritanrix-HepB, Hiberix, and Poliorix. The primary endpoint analysis pooled the original groups according to the booster pneumococcal vaccine, rather than presenting the main comparison separately for each country and polio vaccine grouping.

The registry describes the study as observer blind in its summary and lists double masking in its design information. These descriptions indicate efforts to limit knowledge of treatment assignment, but further details about who was masked and how masking was maintained are not reported in the registry record. The allocation sequence and implementation details are also not reported in the registry record.

This posting concerns objectives and outcomes of the booster phase. The primary phase is identified as being in a separate protocol posting, NCT00344318. The registry record does not give enough information to infer the original sample-size justification, detailed randomization procedures, or the full protocol. The enrolled total and the numbers used in the pooled primary analysis are distinct registry figures; the analysis population is specifically described as vaccinated subjects with symptom sheets filled in.

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: During the screening the following steps occurred: check for inclusion/exclusion criteria, contraindications/precautions, medical history of the subjects and signing informed consent forms.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The analysis was performed on the Total Vaccinated Cohort, which included all vaccinated subjects with the symptom sheets filled in. For the purpose of the analysis, subjects were pooled into two groups, according to the booster vaccine they have received (Synflorix™ or Prevenar™).

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 Subjects Reporting Rectal Temperature Greater Than (>) the Cut-off, measured over Within the 4-day (Days 0-3) period after booster vaccination.

The primary endpoint counts participants reporting rectal temperature > 39.0°C after booster vaccination. The assessment window was the 4-day (Days 0-3) period after vaccination. The registry describes the cut-off as 39.0 degree Celsius and identifies the outcome unit as participants, so the endpoint is a count of children meeting the fever definition, not a measurement of average temperature or duration of fever.

A fever count is a binary participant-level outcome for this analysis: each child either met the specified threshold during the window or did not. This kind of endpoint allows a direct comparison of how often the defined event was reported in the two pooled groups. It is one specific safety measure, not a complete account of all symptoms or adverse events. The registry lists additional safety outcomes, including solicited symptoms, unsolicited adverse events, and serious adverse events, as well as antibody outcomes.

As a general statistical point, a prespecified threshold and observation window make the event definition explicit and comparable between groups. The result nevertheless depends on the recorded temperatures and completion of symptom sheets. The registry says that the primary analysis included vaccinated subjects with symptom sheets filled in; how missing or incomplete symptom information was handled beyond that definition is not reported in the registry record.

7. Results Posted in the Registry

• Prevenar Pooled Group: 10.6% (20/189) versus 11.5% (64/558) on control; risk difference -0.9 percentage points (95% CI -6.0 to 4.2), risk ratio 0.92 (0.57–1.48); unadjusted two-proportion z-test p = 0.739.

8. Statistical Interpretation

9. Statistical Methods Explained

The endpoint is binary at the participant level: a child either reported rectal temperature > 39.0°C during the specified period or did not. Counts and group sizes are posted for the two pooled booster-vaccine groups. The registry’s primary population is the Total Vaccinated Cohort, defined as vaccinated subjects with symptom sheets filled in. This definition matters because the endpoint depends on symptom-sheet information as well as vaccination.

What does the recalculated risk difference express?

The recalculated risk difference compares the event proportion in the Prevenar pooled group with the event proportion in the Synflorix pooled group. Its estimate is -0.009, with a 95% confidence interval from -0.060 to 0.042. A negative estimate in this stated comparison means the observed event proportion was lower in the Prevenar group; it does not by itself establish a reliable group difference. The interval includes zero and also includes values on either side of zero.

The associated unadjusted p-value is 0.739. A p-value describes how incompatible the observed data are with a specified null model, under the assumptions of the test. It is not the probability that the null hypothesis is true, nor is it a measure of the size or importance of an effect. The registry record does not describe the model or calculation details behind this recalculated comparison. It should therefore be understood as a simple unadjusted comparison of posted summary statistics, not as a substitute for the sponsor’s analysis.

Why can the analysis populations and comparisons differ?

The primary analysis pooled participants by booster pneumococcal vaccine. The original treatment groups also differed by country, primary vaccination schedule, and polio vaccine. Pooling supports a single overall comparison of the two booster groups, but it can obscure variation across those original groups. Separate subgroup results for the primary fever endpoint are not reported in the registry record.

The registry lists randomized allocation, double masking, and a parallel model. Randomization is generally used to make treatment groups comparable in expectation, while masking can reduce bias in care, measurement, or assessment. Those are general design principles; details of the sequence generation, allocation concealment, masking roles, and endpoint assessment procedures are not reported in the registry record. The registry’s summary calls the study observer blind, while its design information lists double masking.

What is the sponsor-reported method?

The sponsor analysis is identified as a Philips' statistical test and reports a difference in percentage of 0.89, with a 95% confidence interval from -4.82 to 5.59. It is classified in the registry as a non-inferiority analysis. The registry record does not report the non-inferiority margin, the test’s detailed specification, or the analysis population for that reported test. Without those details, the result cannot be fully reconstructed or its non-inferiority criterion independently assessed from this record alone.

The recalculated risk difference and sponsor-reported percentage difference are expressed in different forms and may reflect different contrast conventions or analysis choices. The registry does not provide enough methodological detail to resolve those differences. They should be described separately, not blended into a single estimate. The discrepancy indicator for this record is false, but that does not supply missing methodological detail.

What does over-dispersion mean in general?

General explanation: over-dispersion occurs when observed variability exceeds what a statistical model assumes. It can make standard errors too small if the model’s variability assumptions are not adequate. The primary outcome is a participant-level binary count, and no dispersion estimate is posted for it. Over-dispersion is therefore not a reported feature of this analysis; it is a general issue to consider when choosing models for other outcomes, particularly outcomes whose variability can exceed a simple model’s assumptions.

How should the wider outcome set be understood?

The registry lists 28 outcome measures and one analysis. The listed secondary outcomes include local and general solicited symptoms, unsolicited adverse events, serious adverse events, pneumococcal antibody threshold counts and concentrations, opsonophagocytic activity titers, and anti-PD concentrations. The registry record does not report results for those secondary outcomes in the registry record used for this interpretation. Their presence among the objectives does not establish what their findings were.

General statistical explanation: when multiple outcomes or comparisons are tested, the chance of at least one apparently notable result can increase if no multiplicity strategy is used. Whether adjustments or a hierarchy were planned for this study is not reported in the registry record. The posted primary comparison is unadjusted, and the registry does not describe interim analyses or a detailed statistical analysis plan. No assumptions about those procedures should be inferred from the single reported analysis count.

10. Confidence Intervals and Precision

The recalculated risk difference is -0.009, with a 95% confidence interval from -0.060 to 0.042. This interval is the main precision summary for the simple unadjusted comparison. Because it includes zero and extends to values on both sides of zero, the result does not precisely determine the direction of the group difference. Its width indicates that the point estimate alone gives an incomplete account of uncertainty.

For context, the recalculated risk ratio is 0.923, with an interval from 0.574 to 1.482. The recalculated odds ratio is 0.913, with an interval from 0.537 to 1.554. Both intervals include the value corresponding to no relative difference. These are alternative effect scales for the same posted group counts, not independent findings.

The sponsor-reported estimate is a difference in percentage of 0.89, with a 95% confidence interval from -4.82 to 5.59. It uses the sponsor’s stated method and is reported separately from the recalculated risk difference. The registry record does not give the details needed to fully reconcile the methods and contrast conventions.

General explanation: a confidence interval describes a range of estimates compatible with the observed data and the statistical procedure, under its assumptions. It does not mean that there is a 95% probability that the true effect lies in this particular interval. Precision reflects how narrowly the data constrain an estimate; it does not by itself establish clinical importance. The registry record does not report a threshold for an important difference in this fever outcome.

The observed group sizes were 558 and 189. The interval should be interpreted in the context of these analysis sizes and the event counts, rather than as a guarantee that an unobserved difference is absent. Results for the original country-specific groups are not reported in the registry record, so pooled precision does not describe the precision of any one original group.

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 illustrations based on the observed effect, not the trial’s original design. For this binary outcome, the figures are participants per arm. The 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 in turn to 100% of the observed effect, 75% of the observed effect, or 50% of the observed effect. The trial’s own original sample-size assumptions are not reported in the registry record.

The 80% power column describes the planned number of participants per arm under those assumptions for a design with 80% power to detect the specified effect. The 90% power column gives the corresponding number per arm for a design with 90% power. Power is a property of a planned study under an assumed effect and model; it is not a measure of whether the observed trial result is true.

For 100% of the observed effect, the planning figures are 19,550 participants per arm for 80% power and 26,171 participants per arm for 90% power. For 75% of the observed effect, they are 35,061 and 46,936 participants per arm. For 50% of the observed effect, they are 79,575 and 106,528 participants per arm. Smaller effects require more participants under the stated assumptions, because they are harder to distinguish from random variation.

These are planning illustrations, not observed power, and they should not be read as evidence about the study’s original planning rationale. The registry record does not report that rationale. The calculations use the observed effect as their starting point, so the large illustrative requirements also reflect the small estimated effect and uncertainty in that estimate. A planning curve, where used, should be read as displaying the same relationship between assumed effect size and required sample size, not as a new result about participants in the completed study.

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 = 0.115; treatment proportion p₁ = 0.106; average p̄ = 0.110.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.4429 = 0.8682.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.4429 = 0.3728.
4. n = (0.8682 + 0.3728)² / (-0.0089)² = 19,549.0, rounded up to 19,550 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 <- 0.114695340501792                          # control proportion
p1 <- 0.105820105820106                          # 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(547248)
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 = 19,550 (80% power), 26,171 (90% power)
 75% of observed effect: n per arm = 35,061 (80% power), 46,936 (90% power)
 50% of observed effect: n per arm = 79,575 (80% power), 106,528 (90% power)
power.prop.test: 19549 
Simulated power at n = 19,550 per arm: 0.795 (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. Analyses Reported by the Sponsor

The sponsor’s reported analysis is a non-inferiority analysis using a Philips' statistical test. It reports a difference in percentage of 0.89, with a 95% confidence interval from -4.82 to 5.59. A p-value is not reported for this analysis. The registry record does not report the non-inferiority margin or enough detail about the test and contrast to independently determine how the sponsor applied a non-inferiority criterion.

The recalculated comparison is a simple unadjusted risk difference, estimated as -0.009 with a 95% confidence interval from -0.060 to 0.042 and an unadjusted p-value of 0.739. It is derived from the posted group counts and sizes and is not the sponsor’s analysis. The sponsor’s percentage difference and the recalculated risk difference are presented on different scales and may use different contrast conventions or procedures. The available registry information does not allow those choices to be fully reconciled.

The sponsor-reported result should therefore be understood as the result of the analysis identified in the registry, while the recalculation is a transparent, unadjusted comparison of the posted summaries. Neither should be described as interchangeable with the other. The record does not include enough detail to assess the sponsor’s full statistical plan or the basis for any non-inferiority conclusion.

14. Multiplicity, Missing Data and Analysis Populations

The registry lists 28 outcome measures, including the primary fever count and secondary measures covering local and general symptoms, unsolicited adverse events, serious adverse events, pneumococcal antibody responses, opsonophagocytic activity, and anti-PD concentrations. The number of reported analyses is one. Results for the listed secondary measures are not reported in the registry record used for this interpretation, so conclusions about those outcomes cannot be drawn from the primary fever comparison.

Multiplicity concerns arise generally when a study examines many outcomes, time points, or comparisons. Without an adjustment or a prespecified testing hierarchy, interpreting many separate p-values can increase the chance of at least one result appearing notable by chance. Whether this study used multiplicity control, a hierarchy, or another strategy is not reported in the registry record. The primary recalculation is explicitly unadjusted; no additional adjustment should be assumed.

Missing-data handling is also not fully described. The primary population is the Total Vaccinated Cohort, consisting of vaccinated subjects with symptom sheets filled in. The record does not report how many vaccinated participants lacked completed symptom sheets, whether any endpoint values were imputed, or whether sensitivity analyses addressed missing observations. The cohort definition identifies who could enter this analysis, but it does not describe every step taken with incomplete data.

The registry does not report interim analyses, stopping rules, or the detailed statistical analysis plan. Those items cannot be inferred from the completed status, the number of analyses posted, or the sample-size illustrations. The record also does not report detailed methods for handling protocol deviations, subgroup comparisons, or the sponsor’s non-inferiority margin. These omissions limit independent assessment of how the primary and secondary analyses were planned and interpreted.

General explanation: prespecification helps distinguish planned analyses from exploratory ones, while transparent reporting makes it possible to assess how missing observations and multiple outcomes affect interpretation. The absence of these details in the registry record is a reporting limitation; it is not evidence that a particular method was or was not used.

15. Safety Results

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

Deaths

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

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: gastroenteritis (2.0%), internal injury (1.1%), brain neoplasm (1.1%), acute respiratory distress syndrome (1.1%), febrile convulsion (1.1%).

The serious adverse event figures are reported for the four original treatment groups, not the two pooled groups used in the primary fever analysis. In the Synflorix, Tritanrix-HepB/Hiberix, and Polio Sabin group, 6 of 280 participants were affected by serious adverse events. In the corresponding Prevenar group, 1 of 93 was affected. In the Synflorix, Tritanrix-HepB/Hiberix, and Poliorix group, 14 of 285 were affected. In the corresponding Prevenar group, 5 of 98 were affected.

One death is listed in the Prevenar group co-administered with Tritanrix-HepB/Hiberix and Polio Sabin; no deaths are listed in the other three groups. The registry also lists serious adverse event terms, including gastroenteritis, internal injury, brain neoplasm, acute respiratory distress syndrome, febrile convulsion, and pneumonia. The posted term percentages are not a substitute for the group-level counts, and the registry information does not establish that a listed event was caused by a vaccine.

These figures are descriptive. They do not provide a controlled causal assessment, and the record does not report further details needed to evaluate event timing, relatedness assessments, or individual clinical circumstances. They should be read separately from the primary fever endpoint and without interpreting the counts as evidence that a treatment is safe or unsafe.

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

The posted primary outcome describes the number of children with rectal temperature > 39.0°C during the 4-day (Days 0-3) period after booster vaccination. The observed counts were 64 of 558 in the pooled Synflorix group and 20 of 189 in the pooled Prevenar group. The unadjusted recalculation estimates a risk difference of -0.009 for Prevenar relative to Synflorix, with a 95% confidence interval from -0.060 to 0.042 and p = 0.739.

These data show a small estimated difference on the recalculated scale, but the uncertainty interval includes zero and values in either direction. They do not establish that the groups are identical, and they do not provide a precise determination of the direction or size of any underlying difference. The p-value does not measure the probability that one vaccine is equivalent to the other. The sponsor-reported non-inferiority analysis is a separate result, with its own reported method and percentage scale; the registry does not include the non-inferiority margin required to interpret that analysis fully.

The endpoint is limited to a particular fever threshold, assessment window, and analysis population. It is not a complete measure of overall safety. The record lists other safety outcomes and immunogenicity outcomes, but their results are not reported in the registry record used for this interpretation. Serious adverse event counts are listed for the four original groups and should not be substituted for the pooled fever comparison.

The comparison is pooled across children from the Philippines and Poland, who had different primary vaccination schedules and received different polio vaccines with the booster. The pooled result does not establish a result for each original group separately. It also does not answer questions about outcomes beyond those posted, longer-term effects, or other populations.

General statistical meaning: estimates describe the data collected under a particular study design and analysis. Uncertainty intervals help show the range of effects compatible with those data and assumptions, while study design and reporting determine how far the results can be generalized. This registry record does not provide every detail needed for a complete evaluation of the analyses or all outcomes.

17. Why This Trial Matters Statistically

This trial illustrates why a point estimate and its uncertainty should be interpreted together. The recalculated risk difference is -0.009, but its 95% confidence interval extends from -0.060 to 0.042. The estimate alone might appear close to zero; the interval shows that the posted data do not pin down the direction and magnitude with high precision. A p-value of 0.739 does not convert that uncertainty into proof of no difference.

It also demonstrates that analysis definitions matter. The primary result is for a specific threshold, rectal temperature > 39.0°C, in a specific post-booster window and in a population defined by vaccination and completed symptom sheets. Counts for that endpoint should not be conflated with other safety outcomes or with immune-response measures.

The original study groups included country, primary schedule, and polio-vaccine differences, while the primary endpoint comparison pooled participants by booster pneumococcal vaccine. Pooling gives an overall comparison but can conceal variation across those original groupings. This makes clarity about the population and contrast essential to interpreting the estimate.

Finally, the registry presents both a sponsor-reported non-inferiority analysis and an unadjusted recalculation. Their reported forms and methods differ, and key details such as the non-inferiority margin are absent. Keeping the analyses distinct helps avoid treating a transparent arithmetic comparison as a replacement for the sponsor’s stated test, while still making the uncertainty in the posted summary counts visible.

18. 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, the complete statistical analysis plan, the sponsor test’s detailed specification, or the contrast convention needed to reconcile its percentage difference fully with the unadjusted recalculation. The sponsor result therefore cannot be independently assessed against its intended non-inferiority criterion from the posted information alone.

The primary analysis pooled children from the Philippines and Poland, despite differences in primary vaccination schedules and co-administered polio vaccines. The record does not report primary fever results separately by original group. The pooled comparison consequently cannot show whether the observed pattern was consistent across those groups.

The analysis population required completed symptom sheets. The number of vaccinated participants without complete sheets and the handling of any missing endpoint information are not reported in the registry record. The record also does not give detailed allocation, concealment, masking implementation, or endpoint assessment procedures. Although design information lists double masking and the summary says observer blind, further operational detail is not reported.

The registry lists a broad set of safety and immunogenicity outcomes, but their results are not available in the registry record used. Multiplicity procedures, interim analyses, stopping rules, subgroup plans, and the original sample-size justification are not reported. The planning illustrations use the observed effect and should not be mistaken for evidence about the trial’s original design.

These limitations restrict interpretation to the posted pooled fever counts and the separately reported sponsor estimate. They do not establish that omitted methods or analyses were absent from the full study materials.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Pneumococcal Vaccine Booster Study in Healthy Children 12-18 Mths Old Previously Primed With the Same Vaccines — NCT00547248.
• Registry record last updated 2019-01-14; status last verified 2017-05.
• Results first posted 2019-01-14.
• 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.

22. Record Summary

NCT00547248 is a completed Phase 3 randomized, double-masked, parallel booster study with enrollment of 756. It evaluated Synflorix and Prevenar boosters in children aged 12 to 18 months who had completed a primary pneumococcal vaccination course, with co-administered vaccines varying across the original country groups.

The primary endpoint counted participants with rectal temperature > 39.0°C within the 4-day (Days 0-3) period after booster vaccination. The pooled groups had 64 of 558 events for Synflorix and 20 of 189 for Prevenar. The unadjusted recalculation estimated a risk difference of -0.009, with a 95% confidence interval from -0.060 to 0.042 and p = 0.739. The sponsor-reported non-inferiority analysis reported a difference in percentage of 0.89, with a 95% confidence interval from -4.82 to 5.59; its p-value and non-inferiority margin are not reported.

The record lists additional safety and immunogenicity outcomes, but their results and several methodological details are not reported in the registry record.