Safety Study of IV Peramivir in Hospitalized Subjects With Confirmed or Suspected Influenza: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Safety Study of IV Peramivir in Hospitalized Subjects With Confirmed or Suspected Influenza (NCT00957996), a randomized phase 3 study. The primary outcome is Change From Baseline in Influenza Virus Titer (48 Hours).

2. Clinical Question

The study concerns antiviral activity, safety, and tolerability in hospitalized subjects with confirmed or suspected influenza. The registry lists influenza and symptoms including cough, sore throat, nasal congestion, myalgia, headache, and fatigue as conditions. The posted primary result addresses short-term viral titer, while the listed secondary outcomes include clinical resolution, symptom and fever resolution, resumption of usual activities, hospital discharge, and influenza-related complications. Those clinical outcomes address different aspects of illness from the laboratory endpoint and should not be treated as interchangeable measures.

3. Trial Design

This was a completed phase 3, randomized, open-label, parallel-group study sponsored by BioCryst Pharmaceuticals. The record gives enrollment as 234 and describes hospitalized subjects with confirmed or suspected influenza infection. The study began in 2009-10, with primary completion in 2010-10. It had no masking: participants and study personnel were not described as blinded to assigned treatment. The two groups received active intravenous peramivir regimens, 300 mg twice daily or 600 mg once daily.

Random allocation can help balance known and unknown participant characteristics between groups on average, but it does not remove the limitations of open-label treatment or establish that a difference in an outcome is caused by the regimen. Because both groups received peramivir, the comparison is between those two regimens. The record does not describe a placebo or untreated group, and this comparison alone cannot establish how either regimen compares with no peramivir.

Enrollment is the study-level figure; the primary endpoint’s reported analysis population is narrower. It includes randomized subjects who received at least one dose or infusion and had confirmed influenza A or B by culture, PCR, or serology. The registry also reports exclusion of 83 subjects from that population because of negative baseline titers. The primary endpoint results are reported for 20 subjects in the 300 mg group and 24 subjects in the 600 mg group. These endpoint-specific figures should not be read as the sizes of the randomized groups for every outcome. For example, secondary outcomes and adverse-event summaries use their own reported denominators.

The record does not report the randomization procedure, allocation concealment, a sample-size justification, or a detailed protocol for deviations from assigned treatment. It also does not report an analysis plan for the primary comparison. Those omissions limit assessment of design and analysis choices beyond the information explicitly recorded.

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

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The Intent-to-Treat Infected (ITTI) population included all randomized subjects who received at least 1 dose/infusion of study drug, and had confirmed influenza A or B by culture, PCR, or serology. Overall, 83 subjects were excluded due to negative Baseline titers (log10 TCID50 0.5).

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: The Intent-to-Treat (ITT) population included all randomized subjects.

6. Primary Endpoint

The registry lists the primary outcome as Change From Baseline in Influenza Virus Titer (48 Hours), measured over Baseline and 48 hours.

The primary endpoint is the change from baseline in influenza virus titer at 48 hours, measured in log10 TCID50/mL. TCID50 is a laboratory measure of infectious virus. A change from baseline expresses how the measured titer shifts relative to the participant’s baseline value; on this scale, a negative change represents a lower measured titer than baseline. That interpretation describes the measurement, not a clinical benefit.

The registry describes a time-weighted change calculated for each subject through 48 hours. It used all available data, applied the trapezoidal rule over the time course, and subtracted the baseline value. The trapezoidal rule is a general method for approximating the area under a curve from observed measurements. A time-weighted summary can reflect measurements across the observation period rather than relying on one isolated value, although the exact influence of each observation depends on the data available for that subject.

The result is expressed as a median for each treatment group, with a 95% confidence interval around that median. A median describes the middle of the observed distribution and is less dependent on unusually high or low values than an arithmetic mean. The registry reports the 300 mg group’s median as -1.66, with a 95% confidence interval from -2.32 to -0.61. For the 600 mg group, it reports -1.47, with a 95% confidence interval from -1.89 to -0.75. These are group-specific summaries; the record does not report a direct primary-endpoint contrast between regimens.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does the median summarize?

The primary result uses a median rather than a mean. In general, the median is the central value after observations are ordered: it is a measure of location that is comparatively resistant to extreme observations. A median can be useful when a distribution is skewed or when a small number of unusually large or small values might pull a mean away from the typical observation. The record identifies the median as the summary measure but does not report the distribution of individual viral-titer changes or explain why this summary was selected.

The median does not describe the spread of all individual results. The reported confidence interval concerns uncertainty around the median estimate, not the range of participant outcomes. The registry gives separate 95% confidence intervals for the two group medians. It does not state the method used to calculate those intervals, so their construction cannot be assessed from the record alone.

How is the time-weighted change formed?

The endpoint uses measurements through 48 hours and calculates a time-weighted change for each subject using all available data. The trapezoidal rule is a general numerical method that approximates the area under a curve by joining neighboring observed points with straight lines. A time-weighted summary can account for measurements at multiple times and the intervals between them, rather than treating a single time point as the whole course. The registry says the baseline value was subtracted after the time-weighted calculation.

Using all available observations in a calculation is not the same as fully specifying how missing observations were handled. The record does not describe imputation, assumptions about why a measurement might be absent, or sensitivity analyses for missing data. It also does not state how a subject with only some post-baseline measurements contributed beyond the phrase “all available data.”

What does the confidence interval communicate?

In general, a confidence interval expresses the uncertainty associated with an estimate under a specified statistical procedure. A narrower interval indicates greater precision under that procedure, while a wider interval indicates less precision. The primary endpoint intervals are reported around each group median, not around a difference between groups. They should not be interpreted as participant-level ranges or as direct evidence that the regimens differ or do not differ.

For a comparison, the relevant quantity would ordinarily be an estimated between-group contrast with its uncertainty, alongside the analysis method. The registry does not report that primary contrast or a primary comparison p-value. Thus, the posted group summaries permit description of each group’s observed median, but not a quantified inference about the difference between the treatment schedules.

Why does the analysis population matter?

The primary endpoint uses the Intent-to-Treat Infected population as defined in the registry. It includes randomized subjects who received at least one dose or infusion and had influenza A or B confirmed by culture, PCR, or serology. This is a particular analysis population, not simply every person enrolled. The record reports that 83 subjects were excluded because of negative baseline titers (log10 TCID50 0.5). It reports primary endpoint results for 20 subjects receiving 300 mg twice daily and 24 receiving 600 mg once daily.

Restricting an analysis to participants meeting infection and treatment criteria can make the endpoint more directly relevant to confirmed infection, but it also narrows the population represented by the result. The registry does not explain the implications of the exclusions for treatment comparability, nor does it provide a primary analysis in a broader population alongside the infected-population analysis.

How should the secondary comparisons be read?

For binary outcomes, a risk difference is the difference between the observed proportions in the groups, expressed on the probability scale. In general, a confidence interval around that difference describes uncertainty in the contrast. An interval that includes zero is compatible with no difference under the interval’s method, but also with values in either direction that the interval allows. A p-value is a measure derived under a specified null model; it is not the probability that a treatment works or that the null hypothesis is true.

The posted secondary comparisons are described as unadjusted recalculations from summary statistics, and the record lists no reported analyses for these outcomes. They are not sponsor-reported inferential analyses. They should therefore be distinguished from a prespecified, adjusted analysis and interpreted cautiously. No calculation in this explanation replaces the registry’s own analysis; the primary endpoint has no posted between-group recalculation to interpret.

What do the planning illustrations mean?

The sample-size illustrations are planning calculations based on observed effects, not observed power and not the trial’s original design justification. Their stated assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, a normal approximation, and an effect to detect set to the stated fraction of the observed effect. The record does not report that these assumptions were used to plan the trial. Planning figures indicate what sample sizes a calculation would imply under its assumptions; they do not establish that the current study had adequate power or that an observed result is likely to replicate.

Planning from an observed effect can be especially unstable when the estimate is imprecise. A small observed effect can yield very large illustrative sample sizes, while an effect estimate based on limited data can change substantially in another study. These calculations should be read as conditional illustrations, not as a claim about the trial’s actual design, observed power, or the number of participants needed in every future setting.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists a primary viral-titer endpoint and multiple secondary outcomes. It does not report a strategy for controlling the chance of false-positive findings across those outcomes, such as a hierarchy of tests or an adjustment for multiple comparisons. Without that information, secondary p-values should be regarded as unadjusted and exploratory. A nominal result from one among several outcomes would not, by itself, establish a reliable finding.

The record lists 13 outcome measures and 11 secondary outcomes posted, but it reports no analyses for the secondary outcomes described. The unadjusted comparisons in the posted results are recalculations from summary statistics, not analyses reported by the sponsor. The registry record does not report a prespecified analysis plan, a primary between-group test for the viral-titer endpoint, or interim analyses. Those items should not be inferred from the study’s phase or completion status.

For the primary endpoint, the registry states that the time-weighted change was calculated using all available data through 48 hours. It does not report how missing measurements were handled beyond that description, whether any values were imputed, or whether analyses tested alternative assumptions about missingness. The record also does not report the extent or pattern of missing primary-endpoint measurements. Consequently, it is not possible to assess from the registry how missing data might affect the estimates.

The primary analysis population is the Intent-to-Treat Infected population, restricted to randomized subjects who received at least one dose or infusion and had influenza A or B confirmed by culture, PCR, or serology. The record reports that 83 subjects were excluded because of negative baseline titers. It does not report a broader-population sensitivity analysis or explain whether exclusions differed by randomized group. These population and missing-data details affect how broadly the primary summaries can be interpreted.

The registry does not report a sample-size justification or the assumptions behind the trial’s original design. The posted planning illustrations are separate calculations using observed effects and stated assumptions; they are not evidence of the trial’s original power or analysis plan.

11. 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 2% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: respiratory failure (2.6%), acute respiratory distress syndrome (2.6%), septic shock (2.6%), renal failure acute (2.6%), chronic obstructive pulmonary disease (1.8%).

The registry reports serious adverse events for 21 of 114 participants at risk in the 300 mg group and 26 of 116 participants at risk in the 600 mg group. Other adverse events are reported for 69 of 114 and 65 of 116 participants at risk, respectively. These are descriptive counts with the denominators recorded for each safety group. They should not be interpreted as a causal comparison: the registry does not report an inferential safety analysis, and the study was open-label.

The serious-event terms listed include respiratory failure, acute respiratory distress syndrome, septic shock, and renal failure acute, each at 2.6%; chronic obstructive pulmonary disease and pneumonia bacterial are each listed at 1.8%. The registry does not give a group-specific breakdown for those listed percentages in the registry record, so they cannot be assigned to a particular regimen. Event terms and percentages also do not, on their own, establish that study drug caused an event.

Deaths are not reported in the registry record. The adverse-event counts should be read as event reporting within the stated safety populations, not as a complete assessment of comparative safety. Differences in participant characteristics, exposure, event ascertainment, and the limited detail available may affect interpretation.

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

The primary result describes short-term viral-titer changes in the registry-defined infected population. The median time-weighted changes are negative in both groups, indicating that the measured titers were lower than baseline on average in the sense captured by the group medians. The confidence intervals convey uncertainty around those medians. They do not reveal the distribution of individual responses or establish a clinical effect.

The result does not establish whether 300 mg twice daily or 600 mg once daily is superior. The registry does not report a direct primary-endpoint contrast, its confidence interval, or a p-value. Separate medians cannot replace that comparison. Nor does the result establish how either regimen compares with placebo or no antiviral treatment, because both randomized groups received peramivir.

The endpoint is a laboratory measure through 48 hours. It does not directly measure symptom relief, clinical resolution, hospital discharge, complications, or return to usual activities. Those outcomes are listed separately, and the available secondary results do not provide a reported sponsor analysis that resolves the primary comparison. A change in viral titer should not be translated into a claim about a participant’s overall clinical course without evidence establishing that relationship.

The record also leaves important questions unanswered. It does not report the method for constructing the primary confidence intervals, a missing-data strategy beyond use of available measurements, a multiplicity plan, a sample-size justification, or a detailed primary comparison. The primary endpoint population excludes subjects with negative baseline titers as described in the registry. These limits constrain precision and generalizability of interpretation.

Accordingly, the posted findings support a descriptive statement about median viral-titer change in the analyzed population and the uncertainty around each group estimate. They do not support a definitive claim that one dosing regimen is better, that either regimen is clinically effective compared with no treatment, or that the regimens have equivalent safety. Such conclusions require comparative analyses and clinical evidence not reported in the registry record.

13. Why This Trial Matters Statistically

This trial illustrates why a randomized design and a clear primary endpoint do not automatically make every result a direct treatment comparison. Randomization assigns participants to groups, but the usefulness of the posted result for comparing groups depends on reporting the relevant contrast and its uncertainty. Here, the primary outcome is summarized separately by regimen, without a reported between-group estimate or test. The two group summaries describe their respective populations but do not quantify the primary difference.

It also highlights the distinction between a laboratory endpoint and outcomes that matter directly to a participant’s clinical course. The viral-titer measure is based on repeated observations through 48 hours and summarized as a time-weighted change. That approach carries information about measured viral levels over time, but it does not make the result equivalent to symptom resolution, hospital discharge, or complications. Each outcome has its own definition, population, and uncertainty.

The study’s infected analysis population is another important statistical feature. It is narrower than enrollment and depends on treatment receipt and laboratory confirmation, with exclusions for negative baseline titers reported by the registry. A clear description of who contributes to an estimate is essential to understanding what population the result represents. Different denominators across endpoint and safety summaries should not be conflated.

Finally, this record shows why analysis plans and reporting details matter. Without a stated primary comparative analysis, missing-data approach, multiplicity strategy, and sample-size justification, readers cannot fully evaluate the inferential framework. Planning illustrations based on observed effects cannot fill those gaps or be interpreted as observed power. Statistical interpretation depends not only on point estimates, but also on the comparison defined in advance, the analysis population, the uncertainty method, and transparent reporting.

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 main limitation for interpreting the primary result is that the registry reports medians and confidence intervals by treatment group but no direct between-group primary analysis. The method used to construct the confidence intervals is not reported, and neither a primary contrast nor its p-value is reported. This prevents a formal assessment of whether the regimens differ on the primary endpoint based on the posted results.

The primary analysis population is restricted to randomized subjects who received at least one dose or infusion and had confirmed influenza A or B. The registry says 83 subjects were excluded because of negative baseline titers, but does not report a broader-population analysis or describe the effect of those exclusions on group comparability. The endpoint-specific group sizes are limited, and the record does not provide individual-level distributions or additional measures of variability.

Although the calculation used all available data through 48 hours, the registry does not report how missing measurements were handled beyond that statement. It does not report the amount or pattern of missingness, imputation procedures, or sensitivity analyses. The choice of analysis population and missing-data handling may affect the estimates, but the available record does not allow that influence to be assessed.

The open-label design means treatment assignment was not masked. The registry does not report details about outcome assessment procedures that would clarify the potential effect of awareness of assignment. The study compares two active regimens, not peramivir with placebo or no treatment. It therefore cannot answer that broader comparative question on its own.

For secondary outcomes, the registry lists multiple measures but does not report analyses for those described. The recalculated unadjusted comparisons do not replace sponsor analyses, and no multiplicity strategy is reported. Safety summaries are descriptive, with no reported comparative safety analysis or death count. These are limitations of the registry record, not evidence that unreported analyses were or were not conducted.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Safety Study of IV Peramivir in Hospitalized Subjects With Confirmed or Suspected Influenza — NCT00957996.
• Registry record last updated 2015-02-12; status last verified 2015-01.
• Results first posted 2015-02-12.
• 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 describes NCT00957996, a completed phase 3, open-label, randomized, parallel-group study of intravenous peramivir in hospitalized subjects with confirmed or suspected influenza. The two active regimens were 300 mg twice daily and 600 mg once daily. Enrollment is listed as 234; the study began in 2009-10 and reached primary completion in 2010-10.

The primary endpoint was time-weighted change from baseline in influenza virus titer through 48 hours, summarized by group median with a 95% confidence interval in the Intent-to-Treat Infected population. The registry reports negative median changes for both groups, but no direct primary comparison between them. It also lists clinical and safety outcomes, while the available secondary comparisons are unadjusted recalculations and no sponsor-reported analyses are listed.

Interpretation is limited by the absence of a reported primary contrast, analysis details, missing-data strategy, multiplicity plan, and sample-size justification. The record supports descriptive reporting of the posted estimates, not a definitive claim of comparative efficacy or safety.