Investigational Therapeutics for the Treatment of People With Ebola Virus Disease: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Investigational Therapeutics for the Treatment of People With Ebola Virus Disease (NCT03719586), a randomized phase 2, phase 3 study. The primary outcome is Mortality.

2. Clinical Question

The clinical question concerns outcomes among people with Ebola virus disease receiving treatment in clinical centers. All four registered arm descriptions include optimized standard of care, so the randomized groups differ in the added investigational regimen rather than in whether supportive care is described as part of treatment. The registered regimens are Remdesivir plus optimized standard of care, MAb114 plus optimized standard of care, REGN-EB3 plus optimized standard of care, and ZMapp plus optimized standard of care.

The registry describes the study as randomized and controlled, with a parallel design and no masking. Random assignment is intended to make treatment groups comparable on average, but it does not by itself establish that observed group differences are precise or statistically persuasive. Those judgments require an analysis that accounts for the outcome, the group sizes, and uncertainty. Inferential comparisons are not reported in the registry record, so the posted mortality counts can be described but cannot establish a treatment contrast with stated statistical uncertainty.

3. Trial Design

The study is registered as a phase 2 and phase 3, multicenter, multi-outbreak, randomized, controlled safety and efficacy study. It is listed as completed, with a reported enrollment of 681. The recorded start date is 2018-11-21 and the primary completion date is 2019-09-09. The trial used a parallel-group structure and had no masking.

Four treatment groups appear in the registered arm descriptions. Each combines an investigational drug with optimized standard of care: Remdesivir, MAb114, REGN-EB3, or ZMapp. The ZMapp group is identified as the control arm. The brief description lists intravenous administration for each treatment, with different schedules: ZMapp was described as three administrations three days apart; Remdesivir as daily administration for 10 days; MAb114 as one administration; and REGN-EB3 as one administration. The registry summary also describes isolation in a clinic, supportive care and monitoring, tests, and follow-up visits.

The primary outcome is the number of participants with mortality by day 28. The posted group counts are 93 for Remdesivir, 61 for MAb114, 52 for REGN-EB3, and 84 for ZMapp. The corresponding outcome-group sizes are 175, 174, 155, and 169. These are descriptive results; no inferential analysis is listed in the registry record. The registry lists three secondary outcomes: time in days to first negative Ebola virus RT-PCR in blood, viremia determined by CTnp values on PCR, and incidence of serious adverse events or adverse events. The detailed analysis methods for these outcomes are not reported in the registry record.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Before assignment: Statistical comparisons for efficacy between the three treatment arms and ZMapp as the control arm were limited to those time periods when contemporaneous enrollments could occur. So, for example, comparison between REGN-EB3 recipients and ZMapp recipients were limited to the time period beginning in January 2019 because that is when the REGN-EB3 arm was added as a fourth arm to the study.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: patients randomized to receive one of the 4 investigational treatments

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: as per the Statistical Analysis Plan, efficacy comparisons between each of the three investigational treatments and the ZMapp control arm were restricted to patients for whom contemporaneous randomization to the study arms was possible.

6. Primary Endpoint

The registry lists the primary outcome as Mortality, measured over 28 days.

The primary endpoint is the number of participants with mortality by day 28 among patients randomized to receive one of the four investigational treatments. It is a count outcome: it records how many participants in each group met the registered mortality outcome within the stated follow-up period. The registry does not further specify the mortality definition in the endpoint description, so details such as cause-of-death classification are not reported in the registry record.

Mortality is a clinically consequential outcome in a treatment study of Ebola virus disease, and a fixed follow-up time gives the outcome a defined observation window. Comparing groups on this endpoint can address whether the observed counts differ across randomized regimens, provided the analysis also accounts for the number of participants assessed and quantifies uncertainty. A count alone is not an estimate of the size or precision of a treatment effect. In particular, the number of events should be interpreted alongside its group denominator, and a formal comparison requires an analysis method and reported uncertainty measures.

The registry posts the group sizes and mortality counts, but does not report a two-arm statistical comparison, confidence intervals, or p values for this endpoint. Consequently, the posted results support a descriptive account of the counts, not a precise estimate of the relative treatment effect.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

General statistical explanation: A randomized trial assigns participants to treatment groups using a chance-based process. This can reduce systematic differences between groups at the point of assignment, but it does not ensure that every characteristic is balanced, nor does it replace statistical analysis. The analysis must define which participants are included, how the endpoint is measured, how missing outcome information is handled, and what comparison is being estimated.

In this study, the primary outcome is a count of participants with mortality by day 28. Counts are informative, but they are not directly comparable without considering the group sizes. A count model or an analysis of group-specific risks could be used in general, depending on the analysis plan and the data structure. The registry record does not report which inferential method, if any, was used for the primary endpoint. It would therefore be inappropriate to describe a particular model as the trial's analysis method.

What does a confidence interval convey?

General statistical explanation: A confidence interval expresses the uncertainty around an estimated quantity under a specified statistical procedure. For a treatment comparison, it can indicate which effect sizes remain compatible with the observed data and the assumptions of that procedure. A narrow interval indicates greater precision than a wide interval, while an interval that includes a null value may be compatible with no difference as well as with effects in either direction. Interpretation depends on the effect measure and the design. No confidence interval for the primary mortality comparison is reported in the registry record.

What does a p value answer?

General statistical explanation: A p value is a measure of how unusual data at least as extreme as those observed would be under a specified null hypothesis and analysis model. It is not the probability that the null hypothesis is true, and it does not measure the size or clinical importance of an effect. A p value also cannot substitute for an effect estimate and its uncertainty. No primary-outcome p value is reported in the registry record, so no statistical significance claim can be made from the posted counts.

Why do denominators matter?

General statistical explanation: A group count is partly determined by the number of participants whose outcomes contribute to it. Denominators help put counts in context, but comparing proportions still requires attention to uncertainty and to the analysis population. The registry gives outcome-group sizes for each arm, but does not explain whether these represent all randomized participants, participants with outcome information, or another analysis set. That distinction is not reported in the registry record.

How should a multicenter, multi-outbreak trial be considered?

General statistical explanation: When a trial spans centers or outbreaks, the setting can matter to how results are interpreted and to how an analysis is specified. Outcomes may vary across settings, and an analysis plan may address such structure when appropriate. The study title describes a multicenter, multi-outbreak design, but the registry record does not report whether center or outbreak was included in an analysis, whether treatment effects were examined by setting, or how any such examination was handled. No subgroup conclusion should be inferred without those details and supporting results.

What does over-dispersion mean for count data?

General statistical explanation: Over-dispersion occurs when count data vary more than a statistical model expects. If a model assumes less variation than is present, its uncertainty estimates can be too small, making findings appear more precise than they are. Appropriate model choice depends on the outcome's distribution and the study design. The registry record does not report a count model, a dispersion estimate, or a check for over-dispersion, so there is no basis for saying whether this issue was assessed in the trial analysis.

The registry includes a description of a planning illustration based on an observed effect, with a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. It specifies that the effect to detect is a stated fraction of the observed effect. These are described as planning assumptions, not as the trial's own design justification or as observed power. The registry record does not report a sample-size justification for the trial. No numerical planning estimates or effect fractions are included in the record summarized for the primary comparison, so no additional planning figure can be interpreted.

10. Multiplicity, Missing Data and Analysis Populations

The study has a primary mortality outcome and lists three secondary outcomes: time in days to first negative Ebola virus RT-PCR in blood, CTnp-based viremia, and incidence of serious adverse events or adverse events. Examining multiple outcomes or multiple treatment contrasts can increase the chance of a false-positive finding if each is tested separately without an appropriate strategy. General approaches include identifying a primary comparison in advance, adjusting for multiple testing, or treating secondary findings as exploratory. The registry record does not report a multiplicity strategy, a hierarchy of tests, or which treatment contrasts were intended for formal testing.

The record lists no analyses and does not report inferential comparisons for the posted primary counts. The analysis population beyond the description of patients randomized to receive one of the four investigational treatments is not reported in the registry record. The method for handling participants with missing outcome information, withdrawals, or incomplete follow-up is also not reported. It is therefore not possible to determine from the record whether the posted counts use an intention-to-treat population, a modified intention-to-treat population, or another analysis set.

General statistical interpretation depends on knowing how missing outcomes were addressed. Excluding participants without recorded outcomes can bias a comparison if the likelihood of missingness differs across groups or relates to prognosis. Other strategies require assumptions that should be stated and assessed. No missing-data method or sensitivity analysis is reported in the registry record. The record also does not report interim analyses, stopping rules, or a detailed statistical analysis plan; these details should not be inferred from the randomized design or the posted outcome counts.

11. Safety Results

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

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: death (1.2%).

The adverse-event listing reports serious adverse events affecting 1 participant in the Remdesivir group, 0 in the MAb114 group, 0 in the REGN-EB3 group, and 2 in the ZMapp group. The listed numbers at risk for these summaries are 175, 174, 155, and 169, respectively. The record also lists other adverse events affecting 6, 1, 2, and 6 participants in those groups, respectively.

These are descriptive counts. They do not establish that one regimen is safer than another, and the absence of a listed serious event in a group does not establish an absence of risk. A safety comparison would require clear event definitions, follow-up and ascertainment information, an analysis approach, and uncertainty estimates. Event definitions and inferential safety comparisons are not reported in the registry record. The registry lists deaths by group separately, but the safety listing does not explain how deaths were classified in relation to serious adverse events.

General statistical explanation: adverse-event counts can be difficult to compare when exposure time, event ascertainment, or severity definitions differ. The record does not report exposure-adjusted safety analyses or detailed event-term results beyond the listed information. The reported counts should therefore be read as the posted safety summaries, not as a complete comparative assessment.

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

The randomized design and posted group-specific outcome counts provide a record of how mortality by day 28 was described for the four treatment regimens. The registry identifies Remdesivir, MAb114, REGN-EB3, and ZMapp, each paired with optimized standard of care, and gives the number of participants in each outcome group and the number with mortality. This is useful descriptive information about the registered trial results.

The counts do not, on their own, establish the magnitude or precision of a difference between regimens. No treatment-effect estimate, confidence interval, or p value for the primary outcome is reported in the registry record. Without these, readers cannot judge the statistical uncertainty around a contrast or whether it meets a stated inferential criterion. The counts alone also do not establish that a difference was caused by the investigational drug.

The record does not report the detailed statistical analysis plan, the analysis population for the posted results, or how missing outcome data were handled. It also does not report formal results for the listed secondary outcomes. These omissions limit interpretation of both the efficacy and safety summaries. A randomized design is a strength for comparative inference, but the design label is not a substitute for a transparent analysis and uncertainty reporting.

Thus, the posted results support a factual description of group sizes, mortality counts, and the adverse-event counts listed in the registry. They do not support a definitive claim about whether any regimen improved outcomes, whether one regimen was superior, or whether the observed differences are statistically reliable. Those conclusions require analyses and details that are not reported in the registry record.

13. Why This Trial Matters Statistically

This trial is statistically notable for its randomized, controlled, parallel-group comparison of four investigational Ebola treatment regimens, each paired with optimized standard of care. The registry describes a multicenter, multi-outbreak study and posts primary mortality counts by group. A randomized comparison can be an important design for evaluating differences between treatment groups because assignment is intended to limit systematic allocation bias.

The record also illustrates why outcome counts need context. The mortality results are counts accompanied by group sizes, but no inferential comparison or uncertainty measure is reported. A reader can describe the posted counts, yet cannot quantify the precision of a treatment contrast or distinguish a stable difference from random variation using the registry results alone. Transparent effect estimates and confidence intervals would be needed for that purpose.

The study lists safety and virologic secondary outcomes in addition to mortality. Such a range of outcomes can describe different aspects of a trial, but requires clear reporting about outcome definitions, analysis populations, missing data, and how multiple comparisons are handled. These details are not reported in the registry record. Their absence makes it important to separate the descriptive data that are posted from broader conclusions that would require additional analysis.

The statistical lesson is not that the observed counts should be ignored, but that they answer a limited question: how many participants in each group were recorded with the endpoint. Claims about comparative efficacy, precision, or safety require more than counts. Reporting the analysis methods and uncertainty around the treatment contrasts would be necessary to support those claims.

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 registry record offers limited detail for interpreting the primary outcome statistically. It posts group sizes and mortality counts but does not report an effect estimate, confidence interval, p value, or inferential comparison. As a result, the precision and statistical reliability of differences between groups cannot be assessed from the record.

The endpoint is described as mortality by day 28, but its detailed definition and ascertainment procedures are not reported in the registry record. The record also does not clarify whether the posted group sizes represent all randomized participants or a subset with outcome data. The analysis population and handling of missing outcomes are not reported, limiting assessment of possible bias from incomplete follow-up or exclusions.

Although the study is described as multicenter and multi-outbreak, the registry does not report how center or outbreak was incorporated into the analysis, whether outcomes were examined across settings, or whether any treatment-effect variation was assessed. The statistical analysis plan, multiplicity approach, interim analyses, and sample-size justification are not reported in the registry record. None should be inferred from the design label or the planning assumptions described for an illustration.

The safety results are also limited to listed counts. Detailed event definitions, event timing, exposure-adjusted analyses, and comparative uncertainty are not reported. The record identifies secondary outcomes but does not provide their inferential results. These limitations mean the registry supports descriptive reporting, while leaving key questions about the analyses and the robustness of comparative conclusions unanswered.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Investigational Therapeutics for the Treatment of People With Ebola Virus Disease — NCT03719586.
• Registry record last updated 2021-09-08; status last verified 2020-11.
• Results first posted 2021-09-08.
• 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.

18. Record Summary

NCT03719586 is a completed phase 2 and phase 3 randomized, controlled, unmasked, parallel-group study of investigational therapeutics for people with Ebola virus disease. It is described as multicenter and multi-outbreak, with reported enrollment of 681. The recorded start date is 2018-11-21 and primary completion date is 2019-09-09.

The four registered regimens were Remdesivir, MAb114, REGN-EB3, and ZMapp, each with optimized standard of care. The primary outcome is the number of participants with mortality by day 28. The registry posts group-specific outcome counts and adverse-event summaries, but lists no analyses. Inferential comparisons, uncertainty estimates, detailed analysis methods, and missing-data procedures are not reported in the registry record.