Study to Compare Lefamulin to Moxifloxacin for the Treatment of Adults With Pneumonia: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Study to Compare Lefamulin to Moxifloxacin for the Treatment of Adults With Pneumonia (NCT02813694), a randomized phase 3 study. The primary outcome is Early Clinical Response (ECR).

2. Clinical Question

The clinical question concerns early response after treatment began, rather than a long-term outcome. Early Clinical Response (ECR) was assessed at 96 hours +/- 24 hours after the first dose of study drug. A participant met the definition through survival, improvement in at least 2 signs and symptoms of community-acquired bacterial pneumonia relative to baseline, no worsening of any such sign or symptom, and no use of concomitant antibiotics for treatment of the pneumonia through the assessment.

This is a composite definition: several clinical conditions must be satisfied for ECR. It therefore represents a particular operational definition of early response, not a measure of each symptom separately. The registry also lists Investigator's Assessment of Clinical Response as a secondary outcome. The detailed definition and results for that secondary outcome are not reported in the registry record.

The registry describes the study as a comparison of efficacy and safety in adults with moderate community-acquired bacterial pneumonia. It does not report further eligibility details in the information, nor a separate primary question about a specific subgroup. The result for the primary comparison is not available as a single estimate in the registry record, so the direction and precision of any treatment-group contrast cannot be characterized from the primary outcome entry.

3. Trial Design

This was a phase 3, randomized, double-blind, double-dummy, parallel-group study of oral lefamulin versus oral moxifloxacin for adults with community-acquired bacterial pneumonia. The registry lists enrollment of 738 participants. The study is marked completed. Its recorded start was 2016-08, and primary completion was 2018-01.

In a randomized parallel-group design, participants are assigned to one of the study groups and followed in that assigned group. Random allocation is intended to reduce systematic differences between groups at the point of assignment. It does not guarantee that every characteristic will be balanced, and it does not eliminate uncertainty from chance variation. The registry identifies randomization but does not report the sequence-generation method, allocation concealment procedures, or any stratification factors.

Double-dummy designs are generally used when groups receive treatments with different forms or schedules: each group can receive its active treatment and a matching placebo for the other treatment. This explanation is general; the registry record does not describe the specific dosing schedule or dummy preparations. The record identifies masking as quadruple, but does not explain which roles were masked or how masking was maintained.

The registered arms were oral lefamulin, 600mg, and oral moxifloxacin, 400mg. The primary endpoint was ECR, assessed at 96 hours +/- 24 hours after the first dose. Its analysis population is described as all randomized subjects. The registry does not report further details of adherence monitoring, treatment discontinuation, follow-up procedures, or handling of participants who did not complete the assessment.

The registry lists 3 outcome measures and 5 analyses, but the detailed analysis plan, model specifications, and ordering of analyses are not reported in the registry record. The primary outcome entry is not presented as a single comparison. Consequently, the design and endpoint can be described, but the primary treatment-effect estimate and its inferential uncertainty cannot be reconstructed from the entry.

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: Subjects who met inclusion criteria and did not meet exclusion criteria were randomly assigned to a treatment group. Administration of study drug was expected to occur as soon as possible after the diagnosis of CABP with all Screening/baseline assessments expected to be completed within 24 hours before the first dose of study drug.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Intent to Treat Analysis Set: All randomized subjects

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 Early Clinical Response (ECR), measured over 96 hours +/- 24 hours after first dose of study drug.

The primary endpoint was Early Clinical Response (ECR), recorded as a count of participants. The registry defines ECR at 96 hours +/- 24 hours after the first dose of study drug. A participant had to be alive, show improvement in at least 2 signs and symptoms of community-acquired bacterial pneumonia relative to baseline, have no worsening of any sign or symptom, and receive no concomitant antibiotics for treatment of the pneumonia through the ECR assessment.

A composite endpoint combines multiple criteria into one classification. This can capture a clinically meaningful overall state while also making the result dependent on every component of the definition. For this endpoint, survival, symptom change, absence of worsening, and antibiotic use all contribute to whether a participant is counted as having an early response. A count of participants therefore does not reveal which component was most influential or how each symptom changed on its own.

Early response is used to assess the clinical course soon after treatment begins. That timing can make it useful for describing an early treatment outcome, but it does not by itself describe later recovery or longer-term outcomes. The registry does not report a separate result for each component of ECR, nor does it report the primary comparison as a single estimate. The posted endpoint definition should therefore be distinguished from evidence about how the groups compared.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Statistical interpretation begins by identifying the population, outcome, comparison, and timing. In this study, the primary outcome was ECR, the comparison was randomized assignment to oral lefamulin or oral moxifloxacin, and the assessment was scheduled for 96 hours +/- 24 hours after the first dose. The registry describes the primary population as all randomized subjects. These features identify the target of the primary analysis, but do not give the analysis procedure or its result.

What does a participant count represent?

ECR was recorded as a count of participants. For a binary endpoint, each participant is classified as meeting or not meeting the endpoint definition, and a group count states how many participants met it. The count alone is not an effect estimate comparing groups: interpretation also requires the relevant group denominators and a defined contrast, such as a difference or ratio in response proportions. The registry entry does not report a single primary comparison from which that contrast can be interpreted.

Binary outcomes can be analyzed using different statistical approaches. Some approaches compare proportions directly; others model the probability of the outcome while allowing adjustment for prespecified factors. These methods can yield different estimates or uncertainty measures, particularly when their assumptions and adjustment choices differ. The statistical model, any covariate adjustments, and the method used for the primary analysis are not reported in the registry record. It is therefore not possible to attribute a particular estimator or inferential method to this trial from the available description.

Why do uncertainty and the analysis plan matter?

An estimated contrast is subject to sampling uncertainty because the enrolled participants are one realization of the population being studied. A confidence interval is a general way to express that uncertainty around an estimate; a p-value is a general measure of how incompatible the observed data are with a specified null hypothesis under a statistical model. Neither quantity, alone, describes clinical importance. For a complete interpretation, the estimate, its uncertainty, the outcome definition, and the planned decision criteria need to be considered together. The primary estimate, confidence interval, p-value, and decision criteria are not reported as a single primary comparison in the registry record.

The registry lists 5 analyses and 3 outcome measures. Those counts do not identify which analysis was primary, how the other analyses were ordered, or whether any adjustments were planned for multiple comparisons. The statistical analysis plan and details of any interim analyses are not reported in the registry record. They should not be inferred from the phase of the trial or from the fact that it was randomized.

What does the analysis population tell the reader?

The primary population is described as the Intent to Treat Analysis Set, consisting of all randomized subjects. In general, an intention-to-treat approach retains participants in their randomized groups for analysis, regardless of later treatment changes. It is often used to preserve the advantages of randomization. The record does not detail how this principle was applied to missing ECR assessments, withdrawals, or other post-randomization events. The definition alone does not establish how those cases were classified or whether all randomized participants contributed an observed outcome.

Missing outcomes require explicit handling because participants without an assessment may differ from those with an observed assessment. Possible approaches include analyzing observed outcomes, imputing missing values, or conducting sensitivity analyses under different assumptions. These are general examples, not methods attributed to this trial. The registry does not report the number of missing primary assessments, their reasons, the method used to handle them, or sensitivity analyses. As a result, the effect of missing data on the primary result cannot be assessed.

How should safety counts be read statistically?

Safety entries report participants at risk, participants affected, and deaths for each group. These are descriptive summaries. Counts can be useful for identifying the scale and distribution of recorded events, but do not by themselves quantify uncertainty or establish a causal relationship. A formal comparison would require a clearly specified safety outcome and analysis, as well as attention to event definitions and follow-up. The registry does not report those details in the information. Serious-event terms and percentages are also listed, but the record does not identify an arm for each term in that list.

In general, over-dispersion matters when count data vary more than a chosen statistical model expects. It can make standard uncertainty estimates too small if ignored. That issue is not a basis for assuming a count model in this trial: ECR is described as a participant count, and the registry does not report a model or a dispersion parameter for its analysis. The appropriate interpretation is limited to what the record actually specifies.

What can be inferred from the available information?

The design and endpoint description support a clear account of the intended randomized comparison. They do not supply a primary effect estimate. Because the primary result is not reduced to one comparison, no estimate, interval, or p-value should be reconstructed by treating an unspecified category or time point as the definitive result. Any recalculation from summary statistics would also require relevant reported outcome counts and a clearly defined contrast; those ingredients are not present as a single primary comparison in the registry record. The sponsor's own analysis takes precedence if a discrepancy between a recalculation and that analysis is reported; no such primary comparison is available in the record.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists 3 outcome measures and 5 analyses. These counts indicate that more than one outcome or analysis was recorded, but they do not reveal the planned hierarchy or how results were to be interpreted together. When several outcomes, time points, or comparisons are tested, the chance of at least one apparently notable result can increase. Statistical methods can address multiplicity by defining a hierarchy, adjusting decision thresholds, or designating analyses as exploratory. This is general explanation; the registry record does not report a multiplicity strategy.

The primary endpoint definition specifies an assessment window, but the primary outcome entry is not presented as one comparison and indicates that it has several categories or time points. The registry does not identify which category or time point should be treated as the singular primary result for interpretation. Details of the protocol's endpoint hierarchy and the planned statistical analysis are not reported in the registry record.

The primary population is described as all randomized subjects. This identifies the intended analysis set, but does not explain how incomplete assessments or withdrawals were handled. The number of participants missing an ECR assessment, reasons for missingness, imputation procedures, and sensitivity analyses are not reported in the registry record. Missingness can affect both the estimated outcome and its uncertainty, depending on why outcomes are unavailable and how the analysis handles them. Those effects cannot be evaluated from the description alone.

The registry does not report whether there were interim analyses, stopping rules, or adjustments related to repeated looks at the data. Those details should not be assumed. Without a reported multiplicity plan and a single primary comparison, isolated results across outcomes or analyses should not be treated as definitive evidence about the primary question.

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 2% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: pneumonia (1.1%), acute myocardial infarction (0.5%), lung abscess (0.5%), acute respiratory distress syndrome (0.5%), anaemia (0.3%).

The registry lists serious adverse events for 17 of 368 participants at risk in the lefamulin group and 18 of 368 in the moxifloxacin group. It lists other adverse events for 61 of 368 and 12 of 368, respectively. Deaths are listed as 5 in the lefamulin group and 3 in the moxifloxacin group. These entries are counts, not adjusted treatment effects, and no statistical comparison or uncertainty interval for them is reported in the information.

The registry also lists serious adverse-event terms with rates: Pneumonia, 1.1%; Acute Myocardial Infarction, 0.5%; Lung Abscess, 0.5%; Acute Respiratory Distress Syndrome, 0.5%; Anaemia, 0.3%; and Atrial Fibrillation, 0.3%. The entry does not identify an arm for each listed term in this description, so these rates should not be assigned to either treatment group.

Adverse-event counts describe events recorded during the study; they do not establish that a treatment caused an event. Interpretation also depends on definitions, follow-up, severity, and how events were collected. Those details are not reported in the registry record. The counts should therefore be read as descriptive safety information rather than a complete causal or comparative safety analysis.

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

The record describes a randomized comparison of oral lefamulin and oral moxifloxacin in adults with community-acquired bacterial pneumonia. It defines the primary endpoint as ECR, gives the assessment window, and identifies the primary analysis population as all randomized subjects. It also reports selected adverse-event counts and serious-event terms. These details establish what the study was designed to examine and provide some descriptive information about recorded safety events.

They do not establish the comparative ECR result. The primary outcome is not reduced to a single comparison, and a primary estimate with its uncertainty is not reported in the registry record. The available information therefore cannot show the direction or magnitude of a primary treatment-group difference, how precisely it was estimated, or whether it met any prespecified statistical criterion. The absence of a single reported comparison should not be interpreted as evidence of no difference.

Likewise, the safety counts cannot by themselves establish that either treatment caused more or fewer adverse events. Counts without a reported inferential comparison, full event definitions, and relevant follow-up details are descriptive. The listed serious-event terms and rates should be interpreted within that limitation.

In general, a trial result applies most directly to the population, interventions, outcome definition, and follow-up actually studied. The registry identifies adults with community-acquired bacterial pneumonia and oral study treatments, but does not provide further eligibility detail in the information. Extrapolation to other populations, treatment regimens, or outcomes cannot be assessed from this record alone.

The appropriate conclusion is narrow: the registry documents a completed phase 3 randomized trial and its planned early-response endpoint, but the primary comparative outcome is not available as one interpretable result in the registry record. A conclusion about comparative efficacy or a complete safety profile would require information that is not reported there.

13. Why This Trial Matters Statistically

This trial illustrates why a well-defined endpoint and an interpretable result are separate requirements for statistical understanding. The registry specifies ECR through survival, symptom improvement, no worsening, and no concomitant antibiotics for pneumonia through the assessment. That definition makes clear what counted as a response, but the endpoint definition cannot show how often participants met it or how the randomized groups compared.

The primary analysis population is identified as all randomized subjects, which is important because analysis populations can affect interpretation of randomized comparisons. Yet that label does not answer how missing assessments were handled or whether all randomized participants contributed observed outcomes. Without those details, readers cannot assess the relationship between the intended analysis population and the actual data analyzed.

The record also lists multiple outcome measures and analyses, while the primary outcome is not presented as a single comparison. This highlights the statistical importance of identifying a prespecified primary result and understanding how multiple outcomes or time points were handled. Otherwise, readers lack a clear basis for distinguishing the main confirmatory question from other analyses.

Finally, the safety entries demonstrate the difference between descriptive counts and comparative inference. Counts can summarize recorded events, but they do not alone establish causation or quantify the uncertainty in a treatment contrast. A careful reading separates the information the registry does report from conclusions requiring an estimate, analysis plan, and uncertainty measure that are not reported 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 main limitation for interpretation is that the registry does not present the primary ECR outcome as a single comparison. It indicates that the outcome has several categories or time points and was not reduced to one comparison. Although the endpoint definition and assessment window are given, the primary effect estimate and its uncertainty are not reported as one interpretable result. The direction and precision of the primary comparison therefore cannot be evaluated from the registry record.

The record does not report the statistical model, covariate adjustments, handling of missing primary assessments, or sensitivity analyses. It also does not report the detailed analysis plan, multiplicity strategy, or whether interim analyses occurred. These omissions limit assessment of how the results were intended to be analyzed and how robust an analysis would be to alternative assumptions. They do not show that a particular method was or was not used; those details are simply not reported in the registry record.

Some design details are also absent from the information. The randomization sequence and concealment procedures, masking roles, detailed eligibility criteria, adherence, and follow-up procedures are not reported in the registry record. The registry identifies quadruple masking, but does not explain which participants or study personnel were masked or how that masking was maintained.

Safety information is limited to reported counts and a list of serious-event terms and rates. The record does not provide a full account of definitions, timing, severity, participant-level overlap, or an inferential comparison for those entries. The listed serious-event terms are not assigned to an arm in the information. These limitations mean that the reported safety figures should not be treated as a complete comparative assessment.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Study to Compare Lefamulin to Moxifloxacin for the Treatment of Adults With Pneumonia — NCT02813694.
• Registry record last updated 2019-10-23; status last verified 2019-09.
• Results first posted 2019-10-23.
• 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

NCT02813694 is a completed phase 3, randomized, double-blind, double-dummy, parallel-group study comparing oral lefamulin, 600mg, with oral moxifloxacin, 400mg, in adults with community-acquired bacterial pneumonia. Enrollment was 738. The recorded start was 2016-08, and primary completion was 2018-01.

The primary outcome was ECR at 96 hours +/- 24 hours after the first dose, defined by survival, improvement in at least 2 signs and symptoms, no worsening, and no concomitant antibiotics for pneumonia through assessment. The intended primary population was all randomized subjects. The registry lists an Investigator's Assessment of Clinical Response as a secondary outcome.

The primary outcome is not reported as a single comparison. Selected adverse-event counts and serious-event terms are listed, but a primary effect estimate, uncertainty measure, detailed analysis plan, and missing-data procedures are not reported in the registry record.