Post-authorisation Passive Enhanced Safety Surveillance of Seasonal Influenza Vaccines: Pilot Study in England: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Post-authorisation Passive Enhanced Safety Surveillance of Seasonal Influenza Vaccines: Pilot Study in England (NCT03278067), a registered phase not stated study. The primary outcome is Weekly Incidence Rates of Any Adverse Events of Interest (AEIs) Reported Via Adverse Drug Reaction (ADR) Card, by Vaccine Group.

2. Clinical Question

3. Trial Design

This was a completed, post-authorisation passive enhanced safety surveillance pilot in England during the 2017/18 influenza season. Its stated purpose was to use passive surveillance enhanced with a reporting card system to record adverse events of interest after influenza vaccination. The registry lists 23939 participants enrolled, a study start date of 2017-09-12, and a primary completion date of 2017-11-30.

The listed intervention is enhanced vaccine safety surveillance, rather than an assigned vaccine treatment. The registry describes three vaccinated groups: recipients of Fluarix Tetra, recipients of non-GSK influenza vaccination, and recipients whose vaccine brand was unknown. Participants were volunteers from 10 volunteer GP practices. The arm descriptions state that vaccination occurred between 01 September and 30 November 2017.

Allocation is not stated, and the registry does not describe a randomized comparison. The groups should therefore be read as observational surveillance categories, not as interchangeable treatment and control arms. Differences between group results, if available, would not by themselves establish that vaccine brand caused those differences. Characteristics that might help interpret group comparability are not reported in the registry record.

The primary outcome was framed as weekly incidence rates of any AEIs reported via the ADR card, expressed as the percentage of subjects. The stated follow-up window was within 7 days post vaccination. The denominator was the weekly vaccinated cohort for the week of interest who received the ADR card; the numerator was those in that denominator who reported any AEIs on the card within the stated follow-up window. The registry also lists weekly EHR-based outcomes for several AEI categories. The detailed protocol, recruitment procedures, and further operational information are not reported in the registry record.

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: One study General Practitioner (GP2) did not follow the procedure as per protocol in distributing, handling of the returned cards and the unused Adverse Drug Reaction (ADR) cards, therefore the data from that GP were excluded from the analysis.

Overall Study

Analysis population for the primary outcome, as defined in the registry: Analysis was performed on the Weekly Vaccinated cohort which included all subjects who were vaccinated with a seasonal influenza vaccine during the week of interest and were registered in electronic health record (EHR), using card-based ADR reporting system during the safety follow-up (FU) period (from vaccination up to 7 days post-vaccination).

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: Participant flow data were only collected for the overall group.

6. Primary Endpoint

The registry lists the primary outcome as Weekly Incidence Rates of Any Adverse Events of Interest (AEIs) Reported Via Adverse Drug Reaction (ADR) Card, by Vaccine Group, measured over Within 7 days post vaccination.

The primary endpoint measures the percentage of participants who reported any adverse event of interest on an adverse drug reaction card within 7 days after seasonal influenza vaccination. For each week of interest, the denominator was the Weekly Vaccinated cohort who received the card; the numerator was participants in that denominator who reported any AEIs on it. The registry says vaccination occurred between week 35 and week 48 of 2017, and that rates were to be estimated by vaccine group. The stated measure includes a 95% Confidence Interval.

This endpoint is suited to describing what the card-based surveillance system captured during the short stated follow-up window. A percentage gives the share of eligible card recipients reporting at least one event, rather than a count of events per unit of person-time. Because the endpoint depends on card receipt and reporting, it reflects reports recorded through that system; it is not, by itself, a complete measure of every event that occurred after vaccination.

The registry record does not report the primary endpoint estimates or their confidence intervals. It also does not provide further detail about how an AEI was defined, how card receipt was documented, or how reports were checked. Those details would matter when assessing the endpoint’s interpretation and completeness.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary measure is a proportion: the number of eligible card recipients reporting any AEIs divided by the number of eligible card recipients in the weekly vaccinated cohort, expressed as a percentage. The registry describes the numerator and denominator in those terms. This differs from an incidence rate based on person-time, which would use accumulated time at risk as its denominator. Although the registry uses the phrase “weekly incidence rates,” the stated unit is percentage of subjects.

What does the confidence interval add?

A confidence interval communicates uncertainty around an estimate. In general, an interval around a percentage indicates a range of values reasonably compatible with the data under the method’s assumptions; it is not a guarantee that the underlying value lies within the interval. The registry specifies a 95% Confidence Interval for the primary endpoint, but the interval estimates are not reported. Without those estimates, the precision of the primary results cannot be judged.

For this endpoint, the denominator is restricted to participants who received the ADR card. That definition is important: card receipt determines who can contribute to the card-based numerator and denominator. In general, if people without cards differ in their likelihood of experiencing or reporting events, a card-based percentage may not represent all vaccinated participants. The registry does not report card receipt totals, how non-receipt was handled, or whether missing cards were assessed.

How should weekly estimates be understood?

A weekly estimate describes the specified cohort for that week, rather than necessarily summarizing all participants across the surveillance period. In general, denominators can vary across weeks, so estimates may have different precision. The record defines the weekly vaccinated cohort and reports that vaccination occurred between week 35 and week 48 of 2017, but does not report weekly denominators or endpoint estimates. It is therefore not possible to determine how the weekly results changed or how much information contributed to each estimate.

The registry also lists EHR-based weekly outcomes covering any AEIs and several categories: fever or pyrexia, local symptoms, general non-specific symptoms, respiratory or miscellaneous events, gastrointestinal events, sensitivity or anaphylaxis, and rash. Those outcomes use EHR records, whereas the primary endpoint uses ADR-card reports. In general, different recording routes can capture different events and depend on different reporting processes. The registry record does not explain how the two sources were reconciled, whether events could appear in both, or whether their definitions were aligned.

Why can group comparisons be difficult?

Random allocation is not stated, and the registry explicitly does not describe a like-for-like control comparison. In general, observational groups may differ in participant characteristics, clinical circumstances, or recording practices. Those differences can influence observed event patterns independently of vaccine brand. The registry does not report an adjustment strategy or enough group characteristics to assess comparability. Consequently, a difference between group estimates, if available, would be descriptive unless a suitable analysis and assumptions supported a stronger interpretation.

Variation among practices can also matter in surveillance data. In general, people within the same practice may share care and recording processes, so their observations may not be fully independent. Over-dispersion is a situation in which observed variation exceeds what a simple statistical model expects; if present, it can make unadjusted uncertainty estimates too narrow. This is a general consideration, not a claim that over-dispersion occurred in this study. The registry does not report whether practice-level variation or over-dispersion was assessed.

What is known about the statistical plan?

The registry reports 132 outcome measures and 0 analyses. It does not report the statistical analysis plan, any model used for the primary endpoint, covariate adjustment, handling of repeated weekly observations, or interim analyses. The 95% Confidence Interval is the stated uncertainty measure, but its calculation method is not reported in the registry record. No significance testing results for the primary endpoint are reported.

Sample-size illustrations are planning aids rather than evidence about the study’s observed results. The stated assumptions for those illustrations are a two-sided significance level of 0.05, equal allocation between two arms, a normal approximation, and an effect to detect expressed as a fraction of the observed effect. The illustrations are not the trial’s own design justification and should not be read as observed power. The registry does not report a sample-size justification for this surveillance pilot. Since the primary effect estimate is not reported, the planning assumptions alone do not establish the study’s ability to detect an effect.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists 132 outcome measures, including the primary card-based endpoint and multiple weekly EHR-based event categories. When many outcomes or group comparisons are examined, the chance of seeing an apparently unusual result by chance can increase. In general, an analysis plan may address this through a prespecified hierarchy, adjustment, or a clearly descriptive interpretation. The registry record does not report a multiplicity strategy, a hierarchy among outcomes, or which outcomes were included in any inferential analyses. It lists 0 analyses.

Missing-data handling is also not reported in the registry record. For the primary endpoint, card receipt defines membership in the denominator. The record does not state how missing cards, incomplete cards, or unavailable EHR records were handled, nor does it describe checks for differences between participants with and without recorded outcomes. In general, incomplete reporting can affect a percentage if the probability of being observed is related to the chance of reporting an event.

The stated analysis population is the Weekly Vaccinated cohort for the week of interest, with ADR-card recipients used for the primary endpoint. The available population description does not provide further detail about exclusions or follow-up completion. The registry record does not report an intention-to-treat population, a per-protocol population, or another analysis-population definition. These omissions limit assessment of how participants moved from enrollment to the endpoint analysis.

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 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: accidental fall (0.0%), unknown, non-causal (0.0%), atrial fibrillation (0.0%), cancer (0.0%), chest pain (0.0%).

The safety listing reports counts of serious events, other events, and deaths for three vaccinated groups. For Fluarix Tetra, the reported figures are 25 serious events among 16433 at risk, 1514 other events among 16433 at risk, and 3 deaths. For non-GSK vaccination, the figures are 3 serious events among 3310 at risk, 133 other events among 3310 at risk, and 0 deaths. For unknown brand, the figures are 10 serious events among 4196 at risk, 66 other events among 4196 at risk, and 0 deaths.

These counts are not the primary card-based endpoint estimates, and they should not be treated as evidence that vaccination caused the events. The registry record does not explain the event ascertainment window for this listing, how seriousness was defined, or whether events were judged related to vaccination. It also does not provide uncertainty intervals for these counts. The listed serious terms include Accidental fall, Unknown, non-causal, Atrial Fibrillation, Cancer, Chest Pain, and Abdominal Pain, each displayed as 0.0%. That display does not supply additional information about the primary endpoint.

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

The registry describes a completed pilot intended to strengthen passive safety surveillance through a card-based reporting system. It defines a primary measure as the weekly percentage of card recipients who reported any AEIs within 7 days after vaccination. That definition clarifies what the study sought to measure and how the denominator was formed.

The registry record does not report the primary endpoint results. It therefore does not establish the observed percentage of card recipients reporting AEIs, the precision of any weekly estimate, or whether the estimates differed across vaccine groups. The status “COMPLETED” indicates the study is listed as completed, but it does not substitute for those outcome estimates.

The listed serious and other event counts provide a record of events by vaccinated group, but they do not establish causal relationships. The groups were not described as randomized, allocation is not stated, and the record says the arms are not a like-for-like control comparison. Observed differences, without adequate adjustment and supporting assumptions, cannot be attributed to vaccine brand.

The surveillance endpoint also reflects the reporting mechanism. An ADR-card measure captures reports made through that system among participants who received a card; it is not necessarily a count of all events that occurred. The record does not report completeness of card return, event validation procedures, or how card reports compared with EHR records.

Accordingly, the registry supports understanding the pilot’s surveillance design and the recorded safety counts, but it does not support a quantitative conclusion about the primary AEI percentage or a causal comparison among vaccine groups. More detailed results and analysis information would be needed to assess those questions.

13. Why This Trial Matters Statistically

This study is statistically relevant because it illustrates how a safety-surveillance endpoint depends on both the event definition and the route through which events are recorded. The primary measure is not simply all events among all vaccinated people: it is the percentage among the specified weekly vaccinated cohort who received an ADR card and reported an AEI within 7 days. Such denominator choices shape the interpretation of any reported percentage.

The pilot also places card-based reports alongside EHR-based weekly outcomes. Different recording systems can have different coverage and reporting behavior. In general, considering these data sources separately can help clarify what each surveillance method captures, while combining them requires clear definitions and methods. The registry does not report how the sources were compared or reconciled.

Its observational group structure is another useful statistical distinction. The vaccine categories describe who received each brand, but allocation is not stated and the groups are not a like-for-like control comparison. Descriptive safety monitoring can identify patterns for further investigation, but group differences in such data are vulnerable to confounding and recording differences. Statistical interpretation must reflect that design rather than treating the groups as randomized arms.

Finally, the record demonstrates why a completed study and a defined endpoint do not alone permit evaluation of findings. The primary estimates, their confidence intervals, and the analysis methods are not reported. Without them, neither event frequency nor precision can be assessed for the primary measure. The registry’s outcome and safety information is therefore most useful for understanding the surveillance framework and its limits, not for making a causal comparison.

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 principal limitation in the registry record is that the primary endpoint estimates and their 95% Confidence Intervals are not reported. Without those results, the central card-based surveillance measure cannot be quantified or assessed for precision. The record also lists 0 analyses and does not provide the statistical analysis plan, model specification, or inferential procedures.

Participants were volunteers at 10 volunteer GP practices, and the registry does not report the characteristics of participants or practices. The extent to which this pilot reflects other settings or populations cannot be assessed from the record. Allocation is not stated, and the vaccine groups are not a like-for-like control comparison. Differences among them may reflect factors other than vaccine brand.

The primary endpoint depends on receiving and completing an ADR card. Card distribution and return totals, missing-card handling, and the completeness of event reporting are not reported. The record also does not specify how AEIs were defined or validated. These gaps limit assessment of potential under-reporting and the consistency of outcome classification.

Although the registry lists multiple EHR-based outcomes and safety counts, it does not report how these sources and measures relate to one another, whether repeated observations were handled, or whether practice-level variation was considered. Multiplicity procedures and interim analyses are not reported. The available safety counts also lack reported uncertainty estimates and details on event ascertainment or causal assessment. These limitations are specific to the information available in the registry record and constrain interpretation of the pilot’s findings.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Post-authorisation Passive Enhanced Safety Surveillance of Seasonal Influenza Vaccines: Pilot Study in England — NCT03278067.
• Registry record last updated 2020-01-13; status last verified 2020-01.
• Results first posted 2020-01-13.
• 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

NCT03278067 is a completed post-authorisation passive enhanced safety surveillance pilot in England during the 2017/18 influenza season, sponsored by GlaxoSmithKline. The registry lists enrollment of 23939, with a study start date of 2017-09-12 and primary completion date of 2017-11-30.

The primary endpoint was the weekly percentage of participants in the vaccinated cohort who received an ADR card and reported any AEIs within 7 days after vaccination. The registry also lists weekly EHR-based event categories and safety counts by vaccine group. It reports 132 outcome measures and 0 analyses. The primary endpoint results, confidence intervals, statistical analysis plan, and details of missing-data handling are not reported in the registry record. The groups were not described as randomized or like-for-like controls, so the record does not establish causal differences by vaccine group.