Impact of Prevnar 13 on Ear Infections in Children: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Impact of Prevnar 13 on Ear Infections in Children (NCT01272999), a registered phase not stated study. The primary outcome is Percentage of Streptococcus Pneumoniae Serotype Isolates From Middle Ear Fluid (MEF) Sample at Surveillance Year 1.

2. Clinical Question

This is a descriptive question about the serotype distribution among evaluable isolates, not a direct comparison of the risk of ear infection between treatment groups. Its focus is the composition of isolates collected from the defined population. The record describes the study as a postmarketing observational study intended to examine the impact of Prevnar 13 on otitis media in children, but the primary outcome is specifically the distribution of serotypes in middle ear fluid samples. The registry does not report a causal estimate of vaccination’s effect on ear infections.

3. Trial Design

The registry describes a completed postmarketing observational study of the impact of Prevnar 13 on otitis media in children. It lists an enrollment of 391 and records the study start as 2010-05-21 and primary completion as 2013-12-19. The sponsor is Pfizer. The listed conditions are acute otitis media and mastoiditis.

The intervention entry is “No intervention,” and the arm description is “Otitis media cases.” The record does not describe randomized allocation, a control group, or a like-for-like comparison. It does not explain how children entered the study, how cases were identified, how surveillance was conducted, or whether participants had received Prevnar 13. Those details are not reported in the registry record, so the study design cannot be reconstructed beyond its observational designation and the listed outcome definitions.

The primary outcome concerns the percentage of Streptococcus pneumoniae serotype isolates from middle ear fluid samples during Surveillance Year 1. The description separates serotypes included in Prevnar 13 from those not included and also accounts for total isolates, clumps, and missing or non-viable isolates. The middle ear fluid population is defined as participants with at least one such isolate. The record cautions that the actual number of participants analyzed for the outcome is unknown because the data were not calculated according to the planned analysis.

The record lists 20 outcome measures and zero analyses. It lists secondary outcomes concerning antimicrobial susceptibility of Streptococcus pneumoniae isolates for several antibiotics in middle ear fluid, as well as ceftriaxone susceptibility in the mastoid population. Completion status does not establish that outcome estimates were posted. The registry record does not include observed primary outcome values, confidence interval limits, or inferential results.

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: This study is a 3-year prospective laboratory-based surveillance study wherein the results of the samples obtained were summarized separately for each year. In this study, no more than 1 sample was obtained from a participant and no more than 1 serotype was isolated from any sample.

Overall Study

Analysis population for the primary outcome, as defined in the registry: MEF population included all participants who have at least 1 S pneumoniae isolate from MEF samples. N (number of participants analyzed) includes total number of participants in MEF population, however actual 'N' for this OM is unknown as this data were not calculated as per planned analysis.

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.

Population: All participants from whom microbiologic specimen of S pneumoniae isolates from MEF or mastoiditis sample were obtained during surveillance period.

6. Primary Endpoint

The registry lists the primary outcome as Percentage of Streptococcus Pneumoniae Serotype Isolates From Middle Ear Fluid (MEF) Sample at Surveillance Year 1, measured over Surveillance Year 1.

The primary endpoint is the percentage of Streptococcus pneumoniae isolates from middle ear fluid samples that fall into specified serotype categories during Surveillance Year 1. The description distinguishes serotypes included in Prevnar 13—3, 7F, 19A, and 19F—from those not included. It also describes calculations for total isolates, clumps, and missing or non-viable isolates. Missing or non-viable isolates include those not sent from an external laboratory, lost or found non-viable during culture before sending, or found non-viable on receipt at the central laboratory.

This endpoint describes the composition of the isolates collected, rather than the frequency of ear infections among all children. In general, reporting percentages by serotype category can characterize which types were represented in the sampled middle ear fluid isolates. It can also make explicit how many isolates could not contribute to a serotype classification, when those counts are reported. The endpoint description specifies an exact two-sided 95% confidence interval based on the observed proportion. A confidence interval expresses uncertainty around an estimated proportion; it does not itself establish a causal effect.

The registry record does not report the observed percentages, interval limits, or the actual number of participants analyzed for this outcome. It says that the participant number for the outcome was unknown because the data were not calculated according to the planned analysis. The endpoint definition is therefore available, but its numerical result and precision cannot be assessed from the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does the endpoint estimate?

The primary endpoint is a proportion: the share of evaluated Streptococcus pneumoniae isolates assigned to a specified serotype category. The outcome description names categories of serotypes included in Prevnar 13 and serotypes not included. It also refers to total isolates, clumps, and missing or non-viable isolates. In general, a proportion is interpretable only alongside a clear numerator and denominator. Here, the intended denominator is described in terms of isolates analyzed, while the population definition refers to participants with at least one isolate. The record says the actual number of participants analyzed for the outcome is unknown because the planned analysis was not calculated, and it does not report the observed isolate counts.

What does the confidence interval convey?

The endpoint description specifies an exact two-sided 95% confidence interval based on the observed proportion. In general, a confidence interval gives a range of values compatible with the data and the stated procedure, reflecting sampling uncertainty. A wider interval indicates less precision than a narrower interval. An interval does not correct for selection, measurement, or design limitations, and it does not establish that an exposure caused an outcome. Since no estimate or interval limits are reported in the registry record, the precision of the primary endpoint cannot be judged.

Why does the analysis population matter?

The outcome description distinguishes the number of participants in the middle ear fluid population from the number of isolates evaluable for the outcome. These are not necessarily interchangeable quantities: a participant may contribute an isolate, and the endpoint is defined over isolates. In general, analyses should make clear whether the unit being counted is a participant, an infection episode, or an isolate, and how multiple observations from the same participant are handled. The registry record does not state how repeated isolates were treated or provide the actual number analyzed. That uncertainty prevents a reader from checking the denominator or understanding how the observations contributed to the estimated percentage.

How should missing and non-viable isolates be understood?

The record defines missing or non-viable isolates to include samples not sent from an external laboratory, samples lost or non-viable during culture before sending, and samples non-viable upon receipt at the central laboratory. In general, missingness matters when the chance that an observation is unavailable is related to its underlying characteristics. If unavailable isolates have a different serotype distribution from evaluable isolates, the observed proportion may not represent the distribution among all relevant isolates. The registry record does not give counts for these categories or report a sensitivity analysis. It is therefore not possible to quantify their contribution or determine whether they could materially affect the estimate.

What role do over-dispersion and dependence play?

In general, standard proportion methods are simplest when observations are independent and the binomial model adequately describes variation. If multiple isolates come from the same child or otherwise share characteristics, their outcomes may be correlated. Treating correlated observations as independent can make uncertainty appear smaller than it is. Over-dispersion means that the observed variation exceeds what a model assumes; methods that account for it can produce more realistic uncertainty. These are general considerations, not claims about what occurred in this study. The registry record does not describe whether participants contributed multiple isolates, whether dependence was assessed, or whether any method adjusted for it.

What can the planning assumptions tell readers?

The planning illustration is described as using a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation, with the effect to detect expressed as a stated fraction of an observed effect. The record also specifies that non-time-to-event illustrations give participants per arm, while time-to-event illustrations give total events. These are planning conventions, not observed study results. A planning sample size is not observed power and does not show that the study detected an effect. The registry record does not report a single comparison, an observed effect estimate for the primary endpoint, or an analysis supporting such a comparison. Thus the planning assumptions cannot be treated as evidence about the study’s outcome.

What does the absence of analyses imply?

The registry lists zero analyses. General statistical interpretation requires estimates, denominators, uncertainty measures, and a clear account of the analysis population. Without posted estimates, a p-value or interval cannot be evaluated; without a defined comparison, an effect contrast cannot be interpreted. The endpoint description alone tells readers what was intended to be measured and how a confidence interval was specified. It does not reveal the observed serotype distribution or whether the planned analysis was completed. Any additional statistical plan, including adjustment, handling of repeated observations, or interim analysis, is not reported in the registry record.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists 20 outcome measures and zero analyses. The primary endpoint is described as a serotype distribution in middle ear fluid isolates, while secondary outcomes concern antimicrobial susceptibility of Streptococcus pneumoniae isolates for several antibiotics in the middle ear fluid population and ceftriaxone susceptibility in the mastoid population. The record does not report which outcomes were analyzed, whether results were selected for reporting, or whether a multiplicity adjustment was planned.

In general, testing many outcomes can increase the chance of finding an apparently unusual result by chance. Statistical approaches can address that issue, but the appropriate approach depends on the prespecified questions and analysis plan. The registry record does not report a multiplicity strategy, a hierarchy among outcomes, or a statistical analysis plan. With no posted analyses, it is not possible to assess how multiplicity was handled or whether it affected any inference.

Missingness is explicitly part of the primary endpoint definition. The record describes missing or non-viable isolates and explains which laboratory pathways fall into that category. It does not report counts in these categories or analyses addressing their potential effect on the observed distribution. In general, missing data can influence an estimate when the chance of being missing is associated with the outcome of interest; the registry does not provide information to evaluate that possibility.

The defined middle ear fluid population includes participants with at least one Streptococcus pneumoniae isolate. The record says the actual number analyzed for the primary outcome is unknown because the data were not calculated according to the planned analysis. It does not report an analysis population flow, the number of evaluable isolates, or how multiple isolates from a participant were treated. These details are necessary for assessing the denominator and the scope of the reported estimate. The record also does not report interim analyses, missing-data methods, or the broader statistical plan.

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

The registry record documents a completed observational postmarketing study with an enrollment of 391 and a stated primary question about serotype isolates from middle ear fluid during Surveillance Year 1. It specifies categories for serotypes included and not included in Prevnar 13, describes handling categories for total, clumped, missing, or non-viable isolates, and states that an exact two-sided 95% confidence interval was intended. These details explain what the primary outcome was meant to characterize.

They do not show the observed serotype distribution. No primary outcome percentages, confidence interval limits, or analyses are reported in the registry record. The record also says that the actual number of participants analyzed for the primary outcome is unknown because the data were not calculated as planned. As a result, the posted information cannot establish the magnitude or precision of any observed proportion.

The registry’s broad study description refers to the impact of Prevnar 13 on ear infections, but the listed primary endpoint is a distribution of isolates, not a reported comparison of ear infection rates between vaccine groups. The intervention is listed as “No intervention,” no comparator is stated, and the listed arm is “Otitis media cases.” The record therefore does not provide the structure needed to interpret a randomized treatment contrast or an observational effect estimate.

In general, descriptive isolate data can help characterize what was recovered from sampled cases, while causal conclusions require a suitable comparison and sufficient information about group formation, measurement, and analysis. The absence of posted estimates means that even the descriptive question cannot be answered numerically from the registry record. The study’s completion status and enrollment entry are not substitutes for results, and the outcome definition alone cannot show whether vaccination changed infection frequency or serotype composition.

12. Why This Trial Matters Statistically

This study is statistically interesting because its registry description connects a postmarketing question about otitis media with a primary endpoint based on the serotype composition of middle ear fluid isolates. That distinction matters: the distribution of isolates is not the same estimand as the frequency of ear infections in a population. Clear identification of the unit of analysis and denominator is essential when interpreting a proportion.

The record also highlights the difference between a defined outcome and a reported result. It specifies serotype categories, a Surveillance Year 1 timeframe, and an exact two-sided 95% confidence interval method. Yet it reports no primary outcome values or analyses, and states that the actual participant number for the outcome is unknown because the data were not calculated according to the planned analysis. Readers can identify the intended measurement but cannot evaluate its estimate or precision.

Missing and non-viable isolates are explicitly recognized in the outcome definition. This is statistically important because laboratory availability can shape which observations enter a descriptive distribution. General principles require attention to the number of evaluable observations and the reasons observations are unavailable, especially when the endpoint is calculated from isolates rather than all enrolled participants. The registry record does not provide the counts needed to assess that issue.

Finally, the observational designation and absence of a stated comparator limit what the posted record can support about impact. A study can be relevant to surveillance while still lacking the information needed for a causal estimate. The record is therefore useful for understanding the intended question and its data structure, but not for determining a numerical result or treatment effect.

13. 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 central limitation is that the registry record contains no observed primary outcome values or listed analyses. It describes a confidence interval method, but gives no proportion or interval limits to interpret. It also states that the actual number of participants analyzed for the outcome is unknown because the data were not calculated according to the planned analysis. This leaves the primary endpoint’s denominator and precision unclear.

The record does not explain how the 391 enrolled participants relate to the middle ear fluid population or to the isolate-level outcome. Enrollment cannot be treated as the number analyzed for the endpoint. The number of evaluable isolates, the number of participants contributing them, and the handling of multiple isolates from a participant are not reported in the registry record. These omissions restrict assessment of representativeness and statistical dependence.

The study is described as observational, the intervention is listed as “No intervention,” and no comparator is stated. The record does not describe how participants or cases were selected, how vaccination status was established, or how potential differences between groups would be addressed. It therefore does not provide a basis for interpreting a causal comparison of vaccination and ear infection outcomes.

The endpoint definition acknowledges missing or non-viable isolates, but the corresponding counts and any analysis of their consequences are not reported. The record also does not report a statistical analysis plan, multiplicity handling, or methods for repeated observations. These missing details limit evaluation of bias, uncertainty, and the intended scope of inference. No additional protocol rationale or sample-size justification is reported in the registry record.

14. Related Tutorials

15. Related Statistical Calculators

16. Sources

• ClinicalTrials.gov: Impact of Prevnar 13 on Ear Infections in Children — NCT01272999.
• Registry record last updated 2023-05-23; status last verified 2020-10.
• Results first posted 2020-11-10.
• 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.

17. Record Summary

NCT01272999 is a completed postmarketing observational study titled “Impact of Prevnar 13 on Ear Infections in Children,” sponsored by Pfizer. The registry records enrollment of 391, a start date of 2010-05-21, and primary completion on 2013-12-19. The listed conditions are acute otitis media and mastoiditis.

The primary outcome is the percentage of Streptococcus pneumoniae serotype isolates from middle ear fluid samples in Surveillance Year 1, classified by whether their serotypes were included in Prevnar 13. The outcome description specifies an exact two-sided 95% confidence interval and defines missing or non-viable isolates. The middle ear fluid population includes participants with at least one such isolate.

The registry lists “No intervention,” an arm of “Otitis media cases,” 20 outcome measures, and zero analyses. It reports no observed primary outcome estimates or confidence interval limits. The actual participant number analyzed for the primary outcome is stated to be unknown because the data were not calculated as planned. No comparator is reported.