LT F-up Study 16-20 Yrs After Vaccine Dose of Hepatitis B With/Without HBIg in Newborns to HBeAg+ Mothers: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for LT F-up Study 16-20 Yrs After Vaccine Dose of Hepatitis B With/Without HBIg in Newborns to HBeAg+ Mothers (NCT00240526), a non-randomized phase 4 study. The primary outcome is Anti-hepatitis B Surface Antigen (Anti-HBs) Antibody Concentration as Measured by Enzyme-Linked Immunosorbent Assay (ELISA)..

2. Clinical Question

This is a question about antibody concentration over long-term follow-up, rather than a reported comparison of clinical hepatitis B outcomes. Although the study title refers to protective efficacy, the listed primary outcome is an antibody measurement. The registry record does not report whether antibody concentrations were linked to clinical outcomes in this follow-up.

3. Trial Design

The registry describes a Phase 4, parallel-group, non-randomized study with no masking. It lists an enrollment of 79. The study began in 2003-10, had primary completion in 2010-03, and is marked completed. No additional subjects were to be recruited during the long-term follow-up, and no vaccine was to be administered during it.

The four groups differed by whether the earlier Engerix schedule included a dose at Month 60 and whether HBIg was administered concomitantly at birth. The three-dose schedules included vaccine at Months 0, 1, and 6; the four-dose schedules added Month 60. Within each schedule, one group received HBIg at birth and the other did not. These descriptions concern prior treatment, not interventions delivered during the follow-up study.

The non-randomized allocation matters for interpretation. Group membership may reflect factors that are not reported in the registry record, and the groups cannot be treated as exchangeable solely because they are listed together. No allocation procedure, matching strategy, or covariate-adjustment plan is reported in the registry record.

The stated purpose of the long-term follow-up was to evaluate persistence of anti-HBs antibodies up to years 16 through 20 after the first vaccine dose. The primary outcome also includes year 15. The primary outcome population was the Long Term According-to-Protocol cohort for immunogenicity, using subjects with available data at each specified time point. The registry record does not report the number assessed at each time point or the number retained in each group.

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: The baseline characteristics are given for the Year 15 time point. Therefore the number of participants in the Engerix-3D Group is 21, as in the Year 15 participant flow data.

Year 15

Year 16

Year 17

Year 18

Analysis population for the primary outcome, as defined in the registry: Analysis was performed on subjects from the Long Term According-to-Protocol (LT ATP) cohort for immunogenicity on subjects with available data at the specified time-points

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 Anti-hepatitis B Surface Antigen (Anti-HBs) Antibody Concentration as Measured by Enzyme-Linked Immunosorbent Assay (ELISA)., measured over At Years 15, 16, 17, 18, 19 and 20.

The primary outcome is anti-hepatitis B surface antigen antibody concentration, measured in mIU/mL with an enzyme-linked immunosorbent assay. The registry describes the result as a geometric mean concentration (GMC) with a 95% confidence interval at years 15, 16, 17, 18, 19, and 20. The outcome therefore records the level of a measured antibody over time; it is not itself a count of infections or a direct clinical outcome.

As a general statistical concept, a geometric mean is useful for measurements that are positive and vary multiplicatively or are strongly right-skewed. It is commonly calculated by taking the mean on a logarithmic scale and transforming back to the original measurement scale. This can make a central summary less dominated by very large values than an arithmetic mean would be. The registry record does not specify the calculation details used for this endpoint.

The assay changed during follow-up. A commercial ELISA was used at years 15 and 16. From year 17 through year 20, concentrations were tested with a validated in-house assay with a cut-off of 3.3 mIU/mL. This is relevant when interpreting measurements across years: results from different assays may not be directly interchangeable without information about their comparability. Such comparability information is not reported in the registry record.

The observed GMCs and confidence intervals are not reported in the registry record. Consequently, the record does not support a description of the antibody levels, their change over time, or the precision of any year-specific estimate.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a geometric mean describe?

As a general statistical explanation, a geometric mean summarizes measurements after taking logarithms, averaging those transformed values, and converting the average back to the original scale. It is suited to positive measurements when relative differences are more meaningful than absolute differences or when the distribution is strongly skewed. A GMC can be interpreted as a central level on the original measurement scale, but it does not show the full distribution of individual values. The registry record specifies GMCs for anti-HBs concentration but does not report their numerical values or the detailed calculation method.

A geometric mean is not interchangeable with an arithmetic mean. The arithmetic mean summarizes values using their ordinary scale, while the geometric mean gives a summary based on multiplicative relationships. If individual measurements are widely dispersed, the same GMC could accompany different distributions. The registry record does not report other distribution summaries, so the variability among individual subjects cannot be reconstructed from the outcome definition alone.

How should a confidence interval be read?

In general, a confidence interval expresses statistical uncertainty around an estimate under the analysis method’s assumptions. A narrower interval indicates greater precision for that estimate, while a wider interval indicates more uncertainty. The interval is not a range expected to contain every individual measurement, and it is not a direct probability statement about a fixed parameter after the interval has been calculated. The registry specifies 95% confidence intervals for the GMCs, but does not report the interval limits. Precision therefore cannot be assessed for any follow-up year.

Confidence intervals also do not address every source of uncertainty. They do not, by themselves, correct for non-random allocation, selective availability of measurements, or differences between assays. Those issues concern the design, measurement, and analysis process, rather than sampling uncertainty represented by an interval.

Why does the assay change matter?

The registry reports that a commercial ELISA was used at years 15 and 16 and that a validated in-house assay was used from year 17 to year 20. The in-house assay had a cut-off of 3.3 mIU/mL. In general, an assay change can affect comparability across time if the methods differ in calibration, sensitivity, or other measurement properties. A validation statement identifies the assay as validated, but the record does not provide details establishing that its results are directly comparable with the commercial ELISA. It also does not report how values at or below the cut-off were treated in the statistical analysis.

This means that a time pattern, if numerical estimates were available, would need to be interpreted with both biological follow-up and measurement method in view. The registry record does not report a bridging analysis or other adjustment for the assay change. No conclusion about the direction or size of any assay-related effect can be drawn from the record.

What does the analysis population imply?

The specified population is the Long Term According-to-Protocol cohort for immunogenicity, among subjects with available data at the specified time points. In general, an available-data analysis uses measurements that exist for each particular analysis. The number contributing can consequently differ across time points. This makes it important to know the number assessed at each year, patterns of loss to follow-up, and reasons measurements were unavailable. Those details are not reported in the registry record.

Available-data estimates can be informative about observed measurements, but they may not represent all enrolled subjects if availability is related to antibody concentration or other characteristics. This is a general concern, not evidence that such a bias occurred in this study. The record does not describe imputation, weighting, sensitivity analyses, or other missing-data procedures, so the effect of missingness cannot be evaluated.

What can be inferred from the non-randomized design?

As a general principle, random allocation can help balance measured and unmeasured baseline characteristics across groups. In a non-randomized study, observed group differences may reflect treatment history, participant characteristics, or other factors. The four groups represent different earlier schedules, but the record does not explain how subjects entered those groups or report an adjusted comparison. Descriptive differences, if available, would not by themselves establish that one schedule caused a difference in antibody concentration.

The registry’s planning assumptions describe an illustration based on an observed effect, with a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The illustration is described as targeting a stated fraction of the observed effect. These are planning assumptions, not evidence about the study’s original design or its observed power. The record does not report the observed effect needed to interpret a specific illustration, and it does not report a statistical plan for the trial’s outcome analyses. Sample-size illustrations should be understood as planning exercises, not as results or as proof of the ability to detect an effect.

10. Multiplicity, Missing Data and Analysis Populations

The registry record lists six outcome measures and reports no analyses. It does not describe a multiplicity adjustment, a hierarchy of hypotheses, or which outcomes, beyond the designated primary outcome, were treated as confirmatory. The existence of multiple listed outcomes does not establish how many statistical tests were conducted. Without reported analyses and a statistical plan, the effect of multiplicity on any inferential claims cannot be assessed.

For the primary outcome, the stated analysis population is the Long Term According-to-Protocol cohort for immunogenicity, with available data at each specified time point. A subject could have an available measurement at one year but not another; the record does not provide the year-specific numbers assessed or explain why measurements were missing. The number of missing measurements, their distribution across groups, and any reasons for missingness are not reported in the registry record.

The record also does not report whether missing values were imputed, excluded under a defined rule, or addressed through sensitivity analyses. As a general matter, conclusions from available measurements can be affected if availability is related to the outcome or to characteristics associated with the outcome. This is a possible limitation of the analysis approach, not a finding that missingness was informative in this study.

Because group allocation was non-randomized, the registry record’s lack of an analysis plan also leaves unclear whether or how baseline differences were considered. No adjusted estimates or covariate strategy are reported. Taken together, the available record identifies the outcome population and its available-data basis but does not provide enough detail to evaluate multiplicity control, missing-data assumptions, or adjustment for group differences.

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

The record describes a study intended to assess persistence of anti-HBs antibodies over long-term follow-up after earlier hepatitis B vaccine schedules, with or without HBIg at birth. Its primary outcome is antibody concentration, summarized as a GMC with a 95% confidence interval at years 15 through 20. This defines what the study set out to measure, but the numerical outcome estimates are not reported in the registry record.

Accordingly, the record does not show the level of antibody measured at any year, whether estimates changed over time, how uncertain they were, or whether the groups differed. It also does not report clinical hepatitis B outcomes for this follow-up. The study title refers to protective efficacy, but the listed primary outcome is an antibody measurement. An antibody measurement alone should not be described as a demonstrated clinical effect, and no conclusion about clinical protection can be drawn from the information.

Even if group-specific antibody values were reported, the non-randomized allocation would limit causal interpretation. The group schedules differed, and the record does not report a design or analysis that makes the groups equivalent apart from those schedules. Any observed difference would need to be interpreted as an association unless the relevant sources of bias were adequately addressed.

These observations are not medical advice. They describe the scope and limits of the registry record: it specifies a long-term immunogenicity outcome and its measurement approach, while leaving the outcome values and important analytical details unreported.

12. Why This Trial Matters Statistically

This study is statistically notable for the gap between a clearly specified long-term measurement question and the limited results detail in the registry record. The outcome is defined by antibody concentration, a geometric mean summary, and confidence intervals at repeated follow-up years. Yet the actual estimates, interval limits, and year-specific analysis counts are not reported. Readers can identify the intended estimand and measurement schedule, but cannot assess the observed levels or their precision.

The design also illustrates why study structure matters when interpreting group summaries. The four groups reflect different prior vaccine schedules and HBIg status, but the allocation was non-randomized. Listing arms side by side does not make them equivalent comparison groups. Without reported group results, adjustment details, and information about group formation, neither the direction nor the cause of any possible difference can be evaluated.

Finally, the assay transition during follow-up is a practical measurement issue. The change from commercial ELISA at years 15 and 16 to a validated in-house assay from year 17 through year 20 means that comparisons over time require attention to method comparability, not only biological interpretation. The record supplies the assay timing and cut-off but does not report cross-assay comparability analyses.

These features make the study useful for illustrating how endpoint definition, design, measurement, and reporting jointly determine what statistical conclusions are supportable. In this record, those elements establish the question and its constraints, but not the outcome findings.

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 main trial-specific limitation for interpreting results is that the numerical primary outcome estimates are not reported in the registry record. The GMCs, 95% confidence intervals, and numbers contributing at each follow-up year are unavailable. This prevents assessment of both the observed antibody levels and their precision.

Allocation was non-randomized, and the four groups differed in prior vaccine schedule and HBIg at birth. The record does not report how subjects entered groups, whether baseline characteristics were balanced, or whether an adjusted comparison was performed. It therefore does not support a causal interpretation of potential group differences.

The analysis population was restricted to the Long Term According-to-Protocol cohort for immunogenicity with available data at each time point. The record does not report how many measurements were unavailable, why they were unavailable, or how missing data were handled. The impact of missingness on estimates cannot be assessed.

Measurement also changed over time: a commercial ELISA was used at years 15 and 16, followed by a validated in-house assay with a cut-off of 3.3 mIU/mL from year 17 through year 20. Cross-assay comparability and handling of measurements relative to the cut-off are not reported. Finally, although the title mentions protective efficacy, the listed primary endpoint is antibody concentration; clinical outcome results are not reported in the registry record.

14. Related Tutorials

15. Related Statistical Calculators

16. Sources

• ClinicalTrials.gov: LT F-up Study 16-20 Yrs After Vaccine Dose of Hepatitis B With/Without HBIg in Newborns to HBeAg+ Mothers — NCT00240526.
• Registry record last updated 2016-12-07; status last verified 2016-10.
• Results first posted 2010-07-15.
• 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

This completed Phase 4 study was non-randomized, parallel-group, and unmasked. It lists enrollment of 79 and evaluates long-term follow-up after earlier Engerix schedules, with or without HBIg administered at birth. The primary outcome is anti-HBs antibody concentration, summarized as a GMC with a 95% confidence interval at years 15 through 20 among subjects with available data in the Long Term According-to-Protocol cohort for immunogenicity.

The assay was a commercial ELISA at years 15 and 16 and a validated in-house assay with a 3.3 mIU/mL cut-off from year 17 through year 20. The primary outcome values, interval limits, year-specific analysis counts, and details of missing-data handling are not reported in the registry record. No analysis results are reported, so outcome patterns and group differences cannot be assessed.