Safety Study of GSK Biologicals' Rotavirus Vaccine (Rotarix®) Administered to Children Aged <1 Year in the United States: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Safety Study of GSK Biologicals' Rotavirus Vaccine (Rotarix®) Administered to Children Aged <1 Year in the United States (NCT00875641), a registered phase not stated study. The primary outcome is Incidence Rates of Intussusceptions (IS) Among the HRV Cohort and Comparator Cohorts.

2. Clinical Question

3. Trial Design

This was a combined prospective and retrospective observational cohort study of Rotarix safety in infants in the United States. The registry describes use of two existing health insurance databases outside the Vaccine Safety Datalink. It records enrollment of 390659 and a status of COMPLETED. The study start date is 2009-04-20, and the primary completion date is 2016-11-04.

The study included an HRV cohort exposed to Rotarix and two IPV-vaccinated control cohorts. One control cohort was concurrent with Rotarix vaccinees; the other was a recent historical cohort. The prospective component compared outcomes following Rotarix and IPV vaccination in the exposed and concurrent control cohorts. The retrospective component identified outcomes following IPV vaccination in the historical control cohort. The record does not state how participants were allocated, and the registry explicitly notes that the arms are not a like-for-like control comparison.

The intervention is listed as a health insurance database rather than as an assigned treatment. That reflects the observational data source: vaccination and health outcomes were studied in routine-care records, not assigned by the study. The study summary describes monitoring of intussusception, Kawasaki disease, convulsions, hospitalizations due to acute lower respiratory tract infections, and all-cause deaths within 60 days of vaccination. The primary outcome was intussusception incidence; the registry also lists secondary outcomes.

The cohort descriptions specify infants younger than one year, but the complete wording of some eligibility descriptions is not reported in the registry record. Detailed recruitment procedures, cohort sizes as analyzed for the primary outcome, and the full analytic protocol are also not reported.

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: Analysis was performed on study population comprising of birth cohorts affiliated with 2 participating health insurance plans, United Healthcare \& WellPoint, which were eligible for rotavirus vaccination according to routine recommendations. Within this population of infants, HRV, Concurrent Control \& Recent Historical Control cohorts were defined.

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 Incidence Rates of Intussusceptions (IS) Among the HRV Cohort and Comparator Cohorts, measured over 60 days following each vaccination.

The primary endpoint is the incidence rate of intussusception in the HRV cohort and comparator cohorts. The registry defines the outcome measure as incidences per person-month and specifies a 60-day risk period following each vaccination. It states that incidence rates were calculated following the first 2 doses of study vaccine, by dividing the number of events by the sum of person-time in the risk period.

Person-time represents the time participants contribute while at risk for the outcome. The registry defines person-time as the number of days at risk and converts person-days to months by dividing by (365.25/12). It specifies a 95% confidence interval for the incidence rate. The numerical rate estimates and their confidence intervals are not reported in the registry record.

As a general statistical point, an incidence rate uses both the event count and the time observed at risk, which can be useful when follow-up time differs among individuals. It is not the same quantity as the proportion of participants who experience an event. The rate describes occurrence relative to person-time; interpretation also depends on how the risk period, outcomes, and eligible person-time were defined and measured.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary measure is an incidence rate, not a simple percentage. The registry defines it as the number of intussusception events divided by the person-time accumulated in the risk period. Its unit is incidences per person-month, and the stated risk period is 60 days following each vaccination. The description says the calculation follows the first 2 doses of study vaccine. The record explains that days at risk were converted to months by dividing person-days by (365.25/12).

Why use person-time?

In general, person-time allows the denominator to reflect how long participants were observed while at risk. When follow-up duration differs, comparing event counts alone can be misleading because groups may contribute different amounts of observation. A rate relates events to that observed time. It does not, by itself, remove bias from differences between cohorts or show that vaccination caused an outcome.

What does the confidence interval convey?

The registry specifies a 95% confidence interval for the incidence rate. In general, a confidence interval expresses the statistical uncertainty around an estimate under the assumptions of the method used. A narrower interval indicates greater precision, while a wider interval indicates less precision. The interval does not account automatically for every possible source of bias, such as outcome misclassification or systematic differences between observational cohorts. Neither the rate estimates nor their intervals are reported in the registry record, so precision cannot be assessed from the posted results.

What is known about the comparison?

The design has one Rotarix-exposed cohort and two IPV-vaccinated control cohorts, one concurrent and one historical. The prospective component compares Rotarix with concurrent IPV vaccination; the retrospective component examines IPV vaccination in the recent historical cohort. The registry states that allocation is not stated and the arms are not a like-for-like control comparison. In general, observational comparisons can be affected by confounding: cohort membership may be associated with characteristics that also relate to outcomes. The registry record does not report a statistical adjustment strategy or adjusted estimates, so it is not possible to establish from the record how measured or unmeasured differences were addressed.

What analysis details are absent?

The record names the primary measure and describes the person-time calculation, but does not report a statistical model, the method used to construct confidence intervals, or a full prespecified analysis plan. A rate calculation can be presented with different inferential methods, and those methods may rely on different assumptions. The appropriate interpretation therefore cannot be reconstructed from the endpoint definition alone. The registry lists 0 analyses; that registry entry is not a numerical estimate of the outcome and does not establish whether analyses were conducted outside the entered results.

The study summary describes a combined prospective and retrospective design using administrative health databases. In general, such records can support observation across routine-care cohorts, but the validity of a specific outcome estimate depends on how vaccination, outcomes, eligibility, and time at risk are captured. Validation procedures and coding details for this study are not reported in the registry record. Nor are the complete analytic population criteria and handling of participants with incomplete records reported.

How should planning illustrations be read?

The registry describes planning illustrations as calculations based on an observed effect, using a two-sided significance level of 0.05, equal allocation, and a normal approximation. For time-to-event calculations, the description says the total number of events is used; other illustrations are expressed as participants per arm. Such sample-size calculations are planning illustrations, not observed power and not evidence that an effect was found. The registry record does not provide the underlying observed effect in the primary results, and no planning illustration can replace the missing outcome estimates.

In general, over-dispersion occurs when variability in event counts exceeds what a simple model assumes. If a method assumes less variability than the data contain, uncertainty can be understated. The registry record does not report whether over-dispersion was assessed or how it was handled; no particular model or dispersion adjustment should be inferred.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists 9 outcome measures. The primary outcome is intussusception incidence. Listed secondary outcomes include incidence rates of Kawasaki disease, convulsions, acute lower respiratory tract infection hospitalisation, and all-cause mortality. The secondary-outcome wording repeats acute lower respiratory tract infection hospitalisation entries. The record does not explain whether repeated entries represent distinct definitions or a duplication.

When several outcomes are examined, the chance of at least one apparently unusual result can increase. In general, a multiplicity plan may specify which outcomes are confirmatory and whether or how inference is adjusted across them. Such a plan, including any hierarchy or adjustment, is not reported in the registry record. The record also lists 0 analyses and provides no numerical outcome estimates from which to assess selective reporting or distinguish primary from exploratory findings.

Missing-data methods are not reported in the registry record. It does not describe the extent of missing vaccination, outcome, or follow-up information, how incomplete records were treated, or whether any sensitivity analyses were performed. These details matter because the rate denominator depends on observed time at risk, and incomplete observation can affect both the amount and composition of person-time.

The primary population description refers to birth cohorts affiliated with two participating health insurance plans and eligible for rotavirus vaccination according to routine recommendations. The full population definition is not reported in the registry record. The record does not provide a complete account of exclusions, cohort entry, participant flow, or the population included in each outcome analysis. Those omissions limit assessment of how the analyzed cohorts relate to the enrolled population and whether the comparison groups were defined consistently.

11. 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.

The adverse-event entries list the HRV cohort with 57931 participants at risk, the concurrent control cohort with 173384, and the recent historical control cohort with 159344. For serious and other adverse events, the affected-participant entries are 0 in each cohort. Death counts are listed separately as 9, 27, and 33, respectively.

These are registry safety entries, not estimates of the primary intussusception incidence rate. The at-risk figures are cohort counts in the adverse-event listing; they should not be read as event counts. The zero affected entries mean that the registry lists no affected participants in those serious and other adverse-event fields. They do not establish that no events occurred outside the scope or reporting rules of those entries.

The record does not report attribution of the deaths to vaccination, their causes, or the relevant observation period for these adverse-event figures. The death counts cannot therefore be interpreted as vaccine-related deaths or compared causally across cohorts. They also do not replace outcome-specific results for intussusception, Kawasaki disease, convulsions, or hospitalizations.

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

This study addresses an important statistical question: how intussusception incidence following Rotarix vaccination compares with incidence in IPV-vaccinated cohorts during a defined post-vaccination risk period. The registry specifies the outcome, person-time denominator, and 95% confidence interval, and describes prospective and retrospective cohort comparisons using two health insurance databases.

However, the registry record does not report the numerical primary incidence rates, their confidence intervals, or the primary outcome event counts. It therefore does not establish the size, direction, or precision of the observed association. Without those results, no quantitative conclusion about the primary comparison can be drawn from the record.

The observational design also limits causal interpretation. Participants were not randomly allocated, and the registry does not state how they entered the cohorts or report adjusted estimates. As a general matter, differences in participant characteristics, time period, healthcare use, or outcome recording can affect comparisons between observational groups. The registry record does not provide enough detail to evaluate those issues fully.

The adverse-event entries should be interpreted separately: they list zero affected participants in the serious and other event fields for each cohort and report deaths as separate counts, without attribution information. These entries do not establish a causal safety effect. Similarly, study completion and a large enrollment figure do not substitute for outcome estimates or demonstrate a particular level of statistical precision.

The appropriate reading is therefore limited to what the record documents: a planned and completed observational safety study with specified cohorts and outcomes, but without posted numerical primary results. The design and endpoint definitions clarify what was intended to be measured; they do not show what the measured rates were.

13. Why This Trial Matters Statistically

This study is statistically notable because it uses routine-care health insurance data to compare a vaccinated cohort with both a concurrent control cohort and a recent historical control cohort. The structure makes clear that safety evidence can be generated through observational designs, while also highlighting why comparator choice and cohort definition matter. A concurrent comparator and a historical comparator do not have identical sources of potential bias.

The primary endpoint is expressed as incidence per person-month rather than only as a count. That distinction focuses interpretation on events relative to observed time at risk. The registry also specifies a post-vaccination window and describes how person-days were converted to person-months, giving readers a basis for understanding the intended rate calculation.

The record is also a useful reminder that endpoint definitions and study completion are not the same as transparent results. It reports the planned outcome and confidence-interval level, but not the primary rates or intervals. Without those quantities, readers cannot assess effect magnitude or precision. The registry lists 0 analyses, making the gap between the study description and posted inferential results especially important to recognize.

Finally, the comparison is observational, and the registry states that the arms are not a like-for-like control comparison. That makes details of eligibility, outcome ascertainment, confounding control, missing data, and multiplicity important to interpretation. Those details are not fully reported. The study thus illustrates both the value of large administrative cohorts for safety monitoring and the limits of inference when the numerical results and analysis details are unavailable in the registry record.

14. 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 primary limitation for interpreting the posted record is that numerical results for the primary incidence outcome are not reported. There are no posted event counts, incidence-rate estimates, or confidence intervals for intussusception. The record therefore does not support assessment of effect direction, magnitude, or precision.

The study is observational, and allocation is not stated. The registry itself cautions that the arms are not a like-for-like control comparison. One control cohort is concurrent and the other recent historical, so differences in cohort composition or calendar period may affect comparisons. Adjusted estimates and a confounding-control strategy are not reported.

The study draws on two health insurance databases and participating health plans. The record does not report how outcomes were validated, how diagnoses were identified, or how complete vaccination and follow-up information was assessed. The complete eligibility criteria and analysis-population definitions are also not reported. Missing-data handling, sensitivity analyses, and a multiplicity plan are not reported in the registry record.

The safety listing gives affected-participant fields and death counts but does not report death attribution or the relevant timing for those entries. They cannot be interpreted as causal effects or as substitutes for the primary outcome. The registry also lists 0 analyses and does not provide a full statistical analysis plan. Taken together, these omissions constrain what can be concluded from the registry record, regardless of the recorded enrollment or completed status.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Safety Study of GSK Biologicals' Rotavirus Vaccine (Rotarix®) Administered to Children Aged <1 Year in the United States — NCT00875641.
• Registry record last updated 2019-01-25; status last verified 2018-06.
• Results first posted 2019-01-25.
• 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.

18. Record Summary

The registry record describes a completed observational cohort study of Rotarix safety among infants in US health insurance plans. It includes an HRV exposed cohort, a concurrent IPV control cohort, and a recent historical IPV control cohort. The design combines prospective and retrospective components and uses two existing administrative health databases outside the Vaccine Safety Datalink.

The primary outcome is intussusception incidence per person-month during the 60 days following each vaccination, following the first 2 doses of study vaccine. The registry specifies a 95% confidence interval and explains the person-time conversion. It lists 9 outcome measures and 0 analyses.

Enrollment is recorded as 390659; the study started on 2009-04-20, reached primary completion on 2016-11-04, and has COMPLETED status. Numerical primary outcome estimates and confidence intervals are not reported in the registry record. The design is observational, allocation is not stated, and the registry notes that the arms are not a like-for-like control comparison.