Comparative Study of Bacille Calmette Guerin (BCG) Delivery Via Disposable Syringe Jet Injector and Needle & Syringe: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Comparative Study of Bacille Calmette Guerin (BCG) Delivery Via Disposable Syringe Jet Injector and Needle & Syringe (NCT01742364), a randomized na study. The primary outcome is Injection Site Adverse Events (Following Injection).

2. Clinical Question

The practical statistical question is how the recorded injection-site event counts and reported tests compare between delivery methods within each age group. For infants, the recorded counts are 138 for the Bioject ID Pen group and 141 for the needle-and-syringe group. For adults, the corresponding counts are 126 and 146. These are event counts, not a direct estimate of the proportion of participants who experienced an event. The registry does not report a confidence interval or a clearly specified effect measure for these comparisons.

The record identifies a randomized, parallel-group study with single masking. It does not report who was masked, how randomization was carried out, or how the event counts were analyzed at the participant level. Those details matter when interpreting comparisons involving repeated or multiple events per participant. The posted results therefore address a limited part of the study's broader question: they describe injection-site events and report tests for infant and adult groups, but do not provide posted immunogenicity results.

3. Trial Design

This was a randomized clinical trial comparing intradermal BCG delivery by a disposable syringe jet injector with delivery by needle and syringe. The registry labels the allocation randomized, the model parallel, and the masking single. It records an enrollment of 96 and lists the study status as completed. The start is recorded as 2012-12 and primary completion as 2013-12. The phase is listed as NA.

The interventions were the Bioject ID Pen and needle and syringe. The registry describes both as methods for intradermal administration of BCG. Results for the primary outcome are split by age group and delivery method: infants using the Bioject ID Pen, infants using needle and syringe, adults using the Bioject ID Pen, and adults using needle and syringe. The reported group sizes are 33 and 33 for the infant groups, and 15 and 15 for the adult groups.

The stated primary objectives were to compare safety and reactogenicity in adults and infants and to compare specific T-cell immunity in neonates. The listed primary outcome is injection-site adverse events following injection, monitored for 14 weeks. The registry record does not report how randomization was generated or concealed, who was masked, or whether masking was maintained. It also does not report a detailed analysis plan or the methods used to handle repeated events, missing observations, or departures from the assigned intervention.

The registry lists 6 outcome measures and 8 analyses, but does not post secondary results. The record therefore supports a description of the basic design and the posted primary outcome results, but not a full reconstruction of the trial protocol or analysis decisions. A sample-size justification and any interim analyses 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.

Overall Study

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: Infants (n=66) and Adults (n=30)

6. Primary Endpoint

The registry lists the primary outcome as Injection Site Adverse Events (Following Injection), measured over 14 weeks.

The primary outcome counts injection-site adverse events after vaccination. The registry's event list includes redness, swelling, induration, tenderness, ulceration, fluctuation, drainage, laceration, bruising, and scarring. Monitoring was specified for 14 weeks following vaccination. These events are relevant to a comparison of reactogenicity because they describe local responses at the injection site after each delivery method.

The reported outcome is a number of adverse events, not a participant-level risk or a measure of event severity. A participant may contribute more than one event, but the registry record does not state how many participants experienced each event, whether repeated events of the same type were counted separately, or how event duration and recurrence were handled. It also does not provide a breakdown of the posted primary outcome counts by event type.

The posted counts are 138 for infants receiving BCG by the Bioject ID Pen and 141 for infants receiving it by needle and syringe. For adults, the counts are 126 with the Bioject ID Pen and 146 with needle and syringe. These totals describe the events as reported; they do not, by themselves, show how events were distributed across participants or how uncertain the comparison is.

Counting events can be useful when the outcome concerns the overall burden of recorded reactions. However, event counts need careful interpretation if participants can have multiple events, because a simple count does not show whether events cluster within a few participants or occur across many. The registry does not report a participant-level summary, a confidence interval, or a detailed definition of the analysis population for this endpoint.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry labels the study randomized and parallel-group. In a randomized trial, allocation is intended to make assigned groups comparable on average, while a parallel design follows separate groups receiving their assigned interventions. These general properties do not establish how allocation was implemented in this study. The sequence-generation and concealment methods are not reported in the registry record. The masking label is single, but the identity of the masked party is not reported.

What does a Fisher exact test assess?

In general, Fisher exact testing evaluates association in a categorical comparison using the observed arrangement of counts, under a specified null hypothesis. It is often used when data are arranged in categories, such as whether an outcome occurred or did not occur. The registry lists Fisher Exact for the infant and adult comparisons and reports p-values of 0.475 and 0.187, respectively. It does not report the contingency tables or explain how the event-count endpoint was converted into, or represented as, categorical data for these tests.

This distinction matters because the primary outcome is described as a number of events, while a categorical test ordinarily operates on category counts. The registry does not state whether each participant was classified according to event occurrence, whether event types were compared, or whether some other analysis unit was used. Without that information, the reported p-values cannot be independently reproduced from the posted group totals, and the test results should not be interpreted as an effect estimate.

Why do event counts need context?

In general, event counts can include multiple observations from one participant. If that occurs, the events are not necessarily independent: a participant who has one reaction may be more likely to have another recorded reaction, or several event types may be documented for the same response. Analyses that treat every event as unrelated can understate uncertainty when outcomes cluster within participants. Count-data methods may also need to account for over-dispersion, meaning that observed variability exceeds what a simple count model would expect. These are general statistical considerations; the registry record does not say whether repeated events occurred or describe a model that accounts for clustering or over-dispersion.

The group totals and group sizes do not yield a participant-level event risk. Such a risk requires information about how many participants experienced at least one event, not just the total number of events. The registry lists group sizes, but does not provide those participant-level counts for the primary outcome. It also does not give event-specific totals or a distribution of the number of events per participant.

How should p-values and uncertainty be read?

A p-value is calculated under a null hypothesis and the assumptions of a particular test. It does not state the probability that the null hypothesis is true, and it does not quantify the size of an observed effect. Confidence intervals are generally useful because they show a range of values compatible with the data under a specified method. No confidence interval for the primary outcome comparison is reported in the registry record. The absence of an interval means the precision of the comparative estimate cannot be assessed directly from the posted summaries.

The registry's stated study hypothesis concerns comparable immune responses and no significant differences in safety or reactogenicity. A claim of comparability or equivalence generally requires a prespecified margin and an analysis designed to evaluate whether effects fall within that margin. The margin and a formal equivalence analysis are not reported in the registry record. A p-value from a test of difference alone does not establish equivalence.

What do the planning assumptions mean?

General sample-size planning asks how many participants may be needed to detect a chosen effect under specified statistical assumptions. The planning assumptions recorded for this study's illustration are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; the target effect is described as a fraction of the observed effect. Such calculations are planning illustrations, not observed power and not evidence that the trial was designed with those assumptions. The registry record does not report the trial's own sample-size justification. The illustration also cannot substitute for a well-defined effect measure and variance appropriate to this event-count outcome.

For time-to-event outcomes, a common planning approach may focus on the total number of events; for other outcomes, planning often focuses on participants per arm. Those general approaches do not resolve the endpoint-definition and analysis-unit uncertainties in this record. No numerical planning result is stated in the registry record, so no required sample size can be inferred from the assumptions alone.

What information would support reproduction?

Reproduction would require a clear outcome definition, the analysis unit, participant-level event occurrence or event-count distributions, and the exact data used in each comparison. It would also require the prespecified statistical plan, the analysis populations, and information on missing observations. These items are not reported in the registry record. The reported group counts and p-values remain useful as a record of the posted results, but they are not sufficient to reconstruct the analysis or calculate a reliable interval around an effect.

10. Analyses Reported by the Sponsor

The registry reports Fisher exact p-values of 0.475 for infants and 0.187 for adults. It does not report an effect estimate or confidence interval for either comparison. The primary outcome is an event count, and the registry does not explain how the counts were represented in the Fisher exact tests. Consequently, an unadjusted recalculation of a two-arm effect is not estimable from the posted summaries for this count endpoint.

No recalculated comparison is reported for these results, so there is no recalculated effect estimate to set alongside the sponsor's analysis. The registry record does not identify a discrepancy between the posted analysis and a recalculation. Where a discrepancy is identified, the sponsor's own analysis takes precedence; that condition does not apply to the results recorded.

This does not mean the reported tests answer every question about the endpoint. Their interpretation remains limited by the absent effect estimates, confidence intervals, participant-level outcome information, and details of the test inputs. The sponsor-reported p-values are best read as the analyses documented in the registry, not as a substitute for an interpretable estimate of event burden or evidence of equivalence.

11. Multiplicity, Missing Data and Analysis Populations

The registry lists 6 outcome measures and 8 analyses, while the posted secondary outcome count is 0. It reports separate primary-outcome tests for infants and adults. The registry record does not explain whether these tests were prespecified as co-primary comparisons, how the broader set of outcomes and analyses was organized, or whether any adjustment for multiple testing was planned or applied.

Multiplicity matters when several outcomes, groups, or analyses are examined. In general, multiple testing can increase the chance of at least one apparently notable result under a collection of null hypotheses. Adjustments or a prespecified hierarchy can address that concern, but the appropriate method depends on the analysis plan. No multiplicity procedure or testing hierarchy is reported in the registry record. The reported p-values should therefore be interpreted as individual results, without assuming a family-wise adjustment or a specific confirmatory status.

Missing-data handling is also not reported. The registry does not describe whether all randomized participants were included in each analysis, how incomplete follow-up was treated, or whether participants with missing event assessments were excluded or otherwise handled. The primary outcome monitoring period is listed as 14 weeks, but the record does not give completion rates or assessment availability during that period.

The analysis populations are not defined in the registry record. It is not stated whether results follow randomized assignment, include only participants who received vaccination, or use another population. The record also does not describe protocol deviations or how they were handled. No interim analyses are reported in the registry record. These omissions limit assessment of the risk of selective analysis and make it harder to interpret the reported tests as the result of a fully specified plan.

Because no secondary results are posted, the stated neonatal T-cell immunity objective cannot be evaluated. The absence of those results does not reveal whether the outcomes were unmeasured, analyzed but not posted, or unavailable for another reason; that explanation is not reported in the registry record.

12. Safety Results

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

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: pneumonia (6.1%), viral pneumonia (3.0%), neonatal jaundice (3.0%), urinary tract infection (3.0%).

The adverse-event information distinguishes serious adverse events from other adverse events. Among infants assigned to the Bioject ID Pen, the registry records 1 serious adverse event among 33 at risk and 30 with other adverse events among 33 at risk. In the infant needle-and-syringe group, it records 4 serious adverse events among 33 at risk and 27 with other adverse events among 33 at risk. For adults, the serious adverse-event count is 0 among 15 at risk in each delivery group, while other adverse events are recorded for 15 among 15 at risk in each group.

These are separate safety summaries from the primary injection-site event totals. Serious adverse events are not interchangeable with injection-site reactions, and the registry does not establish that the serious events were caused by vaccination or by the delivery device. The record lists pneumonia at 6.1%, viral pneumonia at 3.0%, neonatal jaundice at 3.0%, and urinary tract infection at 3.0% among the top serious adverse event terms. It does not provide the denominators or a detailed relationship between those terms and the group-specific serious-event counts.

Zero recorded serious adverse events in an adult group does not show that the possibility of such events is absent. The posted summaries do not include confidence intervals, event narratives, timing, or attribution. These figures describe what the registry records; they are not a complete causal assessment of safety.

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

The registry documents a randomized comparison of two intradermal BCG delivery methods and reports injection-site event totals separately for infants and adults. It also reports Fisher exact p-values for those age-specific comparisons. These results provide a record of the observed event counts and the tests reported by the sponsor, but they do not provide a complete estimate of the difference in event burden between methods.

The counts do not tell how many participants experienced at least one reaction, how events were distributed across participants, or which event types account for the totals. The registry does not report a confidence interval or a clearly defined effect measure. The p-values do not establish that the methods are equivalent, and the record does not report an equivalence margin or formal equivalence analysis. No clinical conclusion about whether either delivery method is preferable follows from these statistical summaries alone.

The results also do not answer the full set of study objectives. In particular, the stated objective of comparing specific T-cell immunity in neonates is not supported by posted secondary results. The primary outcome addresses injection-site events over 14 weeks, not every aspect of safety, reactogenicity, or immune response. The adverse-event summaries add information about serious and other events, but do not provide detailed attribution or a complete causal analysis.

General statistical interpretation depends on a specified analysis unit and method. Here, the registry does not describe how the event counts were used in the Fisher exact tests, whether multiple events per participant were handled, or how missing data and analysis populations were defined. These gaps limit reproducibility and the precision of any comparative interpretation. They do not invalidate the fact that the registry reports the stated counts and p-values; they define what those results can support.

In short, the posted record is descriptive and incomplete for broader claims. It supports reporting the event totals and the sponsor's tests as recorded. It does not support a reconstructed effect estimate, a determination of equivalence, or a conclusion about unreported immune outcomes.

14. Why This Trial Matters Statistically

This trial raises a useful statistical issue: the outcome is reported as a total number of injection-site adverse events, while the group sizes describe participants. Counts and participant risks answer different questions. A total event count describes the recorded burden of events, whereas a participant-level risk describes how many people experienced at least one event. Without participant-level outcome summaries, one cannot tell whether events were spread across the group or concentrated among some participants.

The record also illustrates why the analysis unit must be clear. When a participant can contribute multiple events, the events may be related rather than independent. A statistical method that ignores such clustering can produce misleading uncertainty. The registry lists Fisher exact tests but does not say how event totals were handled in those tests. That leaves a gap between the outcome's description and the reported test method.

Another important issue is the distinction between testing for a difference and demonstrating comparability. A p-value alone does not show that two methods are sufficiently similar. A formal claim of equivalence normally needs a margin and a design and analysis that directly assess that margin; neither is reported in the registry record. This distinction is especially relevant because the study description refers to a comparable immune response, while no immune results are posted.

Finally, separate infant and adult results, a broader set of listed outcomes and analyses, and absent secondary results make prespecification and multiplicity important to interpretation. The record does not state how those analyses were prioritized or adjusted. The trial is therefore a reminder that numerical results are most informative when outcome definitions, analysis populations, test inputs, and uncertainty estimates are all reported alongside them.

15. 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 limitation for statistical interpretation is that the primary outcome is an event count, but the registry does not provide participant-level event occurrence or the distribution of event counts per participant. It does not report event-specific totals, recurrence rules, or the exact analysis unit. The listed Fisher exact tests therefore cannot be reconstructed from the posted group totals, and the p-values cannot be linked to a clearly stated effect measure.

No confidence intervals or comparative effect estimates are reported for the primary outcome. The absence of these estimates prevents direct assessment of precision and makes it difficult to judge what range of differences is compatible with the data. The registry also does not report an equivalence margin, so the stated aim of comparable outcomes cannot be evaluated as a formal equivalence question.

The adult and infant comparisons are presented separately. The registry does not describe a combined analysis or explain whether age-group comparisons were prespecified in a hierarchy. It also does not report how the 6 listed outcomes and 8 analyses relate to the primary analysis, or whether multiplicity was addressed. No secondary results are posted, including results for the stated neonatal T-cell immunity objective.

Further details absent from the registry record include the randomization procedure, allocation concealment, identity of the masked party, analysis populations, missing-data methods, and handling of protocol deviations. The record does not provide a sample-size justification or interim-analysis information. These limitations constrain reproducibility and make it difficult to assess precision, selective analysis, and the scope of the safety findings. They do not change the reported counts; they limit the conclusions that can be drawn from them.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Comparative Study of Bacille Calmette Guerin (BCG) Delivery Via Disposable Syringe Jet Injector and Needle & Syringe — NCT01742364.
• Registry record last updated 2017-02-08; status last verified 2013-12.
• Results first posted 2017-02-08.
• 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.

19. Record Summary

The registry describes a completed, randomized, single-masked, parallel-group trial comparing intradermal BCG delivery by Bioject ID Pen with delivery by needle and syringe in adults and infants. Enrollment is recorded as 96. The study start is listed as 2012-12 and primary completion as 2013-12.

The primary outcome is injection-site adverse events monitored for 14 weeks. The posted totals are 138 and 141 for the infant groups and 126 and 146 for the adult groups, with Fisher exact p-values of 0.475 and 0.187, respectively. No effect estimates or confidence intervals are reported. Secondary results, including the stated neonatal T-cell immunity objective, are not posted.

The registry does not explain the analysis unit behind the Fisher exact tests, define the analysis populations, or report missing-data or multiplicity methods. The results are therefore best read as recorded event totals and sponsor-reported tests, with limited information for reconstructing comparative effects or assessing their precision.