Trial to Evaluate Efficacy and Safety of Nitazoxanide in the Treatment of Mild or Moderate COVID-19: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Trial to Evaluate Efficacy and Safety of Nitazoxanide in the Treatment of Mild or Moderate COVID-19 (NCT04486313), a randomized phase 3 study. The primary outcome is Time to Sustained Clinical Recovery.

2. Clinical Question

3. Trial Design

This was a Phase 3, randomized, double-blind, placebo-controlled, parallel-group trial sponsored by Romark Laboratories L.C. The condition listed is COVID-19, and the title describes evaluation of nitazoxanide in mild or moderate COVID-19. The registry record marks the study as completed. Its recorded start date is 2020-08-13, and its primary completion date is 2021-02-08. Enrollment is listed as 935.

The nitazoxanide arm is described as two 300 mg tablets taken orally twice daily for 5 days. The placebo arm is described as two placebo tablets taken orally twice daily for 5 days. The record lists triple masking, but does not identify the roles covered by that masking in the registry record. It also lists Vitamin Super B-Complex as an intervention, without describing its use in either arm. No conclusions about its assignment or purpose can be drawn from that listing alone.

The primary endpoint was time to sustained clinical recovery, with a timeframe of up to 21 days. The efficacy population is defined as the ITTI population, consisting of subjects who tested positive for SARS-CoV-2 by RT-PCR at baseline. The reported endpoint summaries include 184 participants in the nitazoxanide group and 194 in the placebo group. Those endpoint-specific group counts should not be confused with the enrollment figure, which is a separate registry entry.

The registry lists one secondary outcome: the proportion of subjects progressing to severe COVID-19. It also records 11 outcome measures and 13 analyses. Details of the allocation procedure, the rationale for enrollment, the masking roles, and the full protocol 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

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: The ITTI (primary efficacy) population consisted of all subjects positive for SARS-CoV-2 by RT-PCR at Baseline.

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 Time to Sustained Clinical Recovery, measured over Up to 21 days.

Time to sustained clinical recovery measures how long it takes for a participant to meet a specified set of symptom and temperature criteria after beginning study medication. The clock starts at the first dose. Recovery is defined by a decrease in the total FLU-PRO score from the previous diary, with the participant reporting that symptoms are at least “somewhat better than yesterday.” The definition also requires no oral temperature at least 100.4 F during the prior 24 hours and no future increase in any FLU-PRO domain except within validated background levels.

This is a time-to-event endpoint: the event is reaching the defined recovery state, and the outcome is the time until that event. Such endpoints can retain information about when recovery occurs, rather than reducing the result to whether recovery happened by a single chosen time. The registry reports medians and an inter-quartile range, with a timeframe of up to 21 days.

The endpoint’s value depends on its operational definition. A participant must meet the diary, symptom, and temperature conditions; general improvement alone is not necessarily the recorded event. The registry’s definition therefore makes clear what counts as sustained clinical recovery for this trial, while the posted summary does not show each participant’s diary history or the timing of individual events.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is a time-to-event measure. In general, time-to-event methods analyze the time until a defined event occurs, such as reaching the trial’s criteria for sustained clinical recovery. This approach can make use of the timing of events. It can also accommodate observations in which the event is not observed during the follow-up period, if the analysis and data support appropriate handling of such observations. The registry entries do not describe how censoring or incomplete follow-up was handled in this trial.

The registry summarizes the endpoint with medians and identifies the dispersion as an inter-quartile range. In general, the median is the middle of an ordered set of observed times. The inter-quartile range describes the spread of the central portion of those observations. These summaries are useful for describing the distribution, but they do not quantify the uncertainty in a between-group effect. The posted arm summaries give the median values and accompanying bounds, but the registry entries do not establish that those bounds are a confidence interval for the difference between arms. They should not be read as such.

The reported comparison uses the Gehan-Wilcoxon test. In general, this is a rank-based method for comparing time-to-event distributions. It assesses evidence of a difference between groups by using event-time information and, depending on its implementation, gives relatively greater weight to earlier events than some other time-to-event tests. A test p-value is not an effect size: it does not tell a reader the magnitude or direction of a treatment effect in the original units. The registry does not report the test statistic, a treatment-effect estimate, or a confidence interval for the primary comparison in the registry record for this analysis.

A hazard ratio is a common effect measure for time-to-event outcomes, but the record’s primary analysis is described as a Gehan-Wilcoxon test and no sponsor-reported hazard ratio is listed in the registry record. A hazard ratio would compare event rates over time under a model; it would not be interchangeable with a difference in median times. No hazard-ratio interpretation should be attached to the posted medians or p-value.

Random allocation is intended, in general, to make treatment groups comparable in expectation with respect to prognostic factors, while masking can reduce the influence of knowledge of assignment on care or outcome assessment. Those are general design purposes, not proof that all sources of bias were eliminated in this trial. The registry identifies randomization, double blinding, placebo control, and triple masking, but detailed procedures and the roles covered by masking are not reported in the registry record.

The registry defines the ITTI primary efficacy population as all subjects positive for SARS-CoV-2 by RT-PCR at baseline. This definition gives a criterion for inclusion in the primary efficacy population; it does not, by itself, describe how missing diaries, withdrawals, protocol deviations, or other post-randomization events were handled. Details of those procedures are not reported in the registry record.

The sponsor’s posted p-value is 0.879. A recalculated two-arm comparison is not estimable from the posted summary statistics for this median time endpoint, so there is no independent recalculated estimate to compare with the sponsor’s test. Recalculation from group medians alone would not reproduce a time-to-event test because the participant-level event-time information and relevant analysis details are not available in the registry entries.

In general, over-dispersion is a concern when the observed variability is greater than a statistical model assumes. For count outcomes, for example, a model that assumes too little variability can produce uncertainty estimates that are too narrow. That general issue should not be conflated with the median time endpoint. The registry entries do not describe an over-dispersion assessment or a model for the secondary outcome.

The sample-size assumptions recorded for planning illustrations describe a two-sided significance level of 0.05, equal allocation, and a normal approximation, with the target effect set as a stated fraction of the observed effect. For time-to-event tables, the stated convention is to report total events. Such figures, when presented, are planning illustrations based on assumptions; they are not observed power and do not replace the trial’s own design rationale. The trial’s own sample-size justification and statistical plan are not reported in the registry record.

Multiplicity arises when multiple outcomes or analyses are considered. In general, testing many hypotheses can increase the chance of at least one apparently notable result unless the analysis plan accounts for the family of tests. The record lists 11 outcome measures and 13 analyses, but the registry entries do not describe a multiplicity adjustment, a prespecified testing hierarchy, or how the primary and secondary results were interpreted together. The reported primary p-value should therefore be understood as the result of its stated test, without assuming a particular adjustment procedure.

10. Analyses Reported by the Sponsor

The registry reports the sponsor’s primary comparison as a Gehan-Wilcoxon test with p = 0.879. The available arm-level medians and accompanying values are not enough to recreate that time-to-event test. As a result, there is no recalculated test result or effect estimate that can be set against the sponsor’s analysis. Recalculating a comparison from the medians alone would not reproduce the test’s use of event-time information.

The registry record does not list a hazard ratio, a confidence interval for a treatment contrast, or the test statistic for this primary analysis in the registry record for this analysis. The reported p-value is therefore the sponsor’s own inferential result, while the arm summaries remain descriptive. The posted data do not show a discrepancy between a sponsor result and a recalculated result; a recalculated comparison is not estimable from the posted summary statistics for this endpoint.

Interpretation should remain anchored to what is actually reported: the median recovery times, the listed accompanying values, and the sponsor’s p-value. The p-value does not supply an estimate of the size or precision of a treatment effect. The absence of a recalculated result does not itself validate the sponsor’s analysis; it means the registry summaries do not permit an independent reproduction of it.

11. Multiplicity, Missing Data and Analysis Populations

The registry lists 11 outcome measures and 13 analyses. The primary outcome is time to sustained clinical recovery. One secondary outcome is listed: the proportion of subjects progressing to severe COVID-19. The number of listed outcomes and analyses makes it important to know which comparisons were prespecified and how the analysis plan treated multiple testing. That information is not reported in the registry record.

In general, examining multiple outcomes or multiple analyses creates opportunities for chance findings. Statistical procedures can address that concern, for example by controlling the error rate across a defined family of tests, but the appropriate approach depends on the prespecified questions and analysis plan. No multiplicity adjustment, testing hierarchy, or family of hypotheses is described in the registry record. The reported primary p-value should not be assumed to be adjusted or unadjusted beyond what the record states.

The primary efficacy population is defined as subjects positive for SARS-CoV-2 by RT-PCR at baseline. That is a population definition, not a full account of missing-data handling. The registry entries do not report how missing symptom diaries, incomplete follow-up, withdrawals, or other missing endpoint information were treated. They also do not explain the relationship between enrollment and the numbers included in the posted endpoint summaries.

The registry record does not report an interim-analysis plan, stopping rules, a detailed statistical analysis plan, or sensitivity analyses for missing data. This absence does not show that such procedures were not used; it means they cannot be assessed from the information available in the registry record. Consequently, readers can describe the named endpoint, population, and reported test, but should avoid assuming unreported procedures or adjustments.

12. 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: covid-19 pneumonia (0.6%), ascites (0.2%), covid-19 (0.2%), diverticulitis (0.2%), syncope (0.2%).

The serious adverse event figures are presented with participants at risk and participants affected. For nitazoxanide, the registry lists 472 at risk for serious adverse events and 2 affected. For placebo, it lists 463 at risk and 7 affected. These are descriptive counts. The record also lists deaths as 2 in the nitazoxanide group and 0 in the placebo group. The registry entries do not state whether the deaths were considered related to study medication.

For other adverse events, 16 participants are listed as affected in the nitazoxanide group and 10 in the placebo group; the recorded at-risk counts are 472 and 463, respectively. These safety summaries should not be treated as formal comparisons unless the relevant definitions, follow-up, and analysis are available. Counts alone do not establish causality or show whether groups differed beyond chance.

The registry lists COVID-19 pneumonia at 0.6% among the top serious adverse event terms. Ascites, COVID-19, diverticulitis, syncope, and aspiration are each listed at 0.2%. These terms and percentages are descriptive registry entries; the registry entries do not provide participant-level details or attribution. Differences in counts should not be interpreted as evidence that a treatment caused or prevented an event.

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

The posted results describe a randomized comparison of nitazoxanide and placebo on time to sustained clinical recovery in the defined primary efficacy population. The median times are 13.3 days for nitazoxanide and 12.4 days for placebo. The sponsor-reported Gehan-Wilcoxon test has p = 0.879. These statistics summarize the endpoint and test; the registry entries do not include a confidence interval or an effect estimate for the between-group comparison.

The result does not establish that the groups are equivalent. Equivalence requires a prespecified margin and an analysis capable of showing that the plausible effect is contained within that margin. Such a margin is not reported in the registry record. A p-value that does not indicate strong evidence against a null model is not proof that there is no difference, nor does it demonstrate that any possible difference is unimportant.

The result also does not establish what happened for each participant. Group medians do not show the full distribution of event times, and the p-value does not describe individual outcomes. The endpoint is specifically defined through diary symptoms and temperature criteria, so it should not be generalized without qualification to other definitions of recovery, other outcomes, or populations not described in the primary efficacy definition.

The record lists progression to severe COVID-19 as a secondary outcome, but the registry entries do not include its result. The posted primary endpoint therefore cannot answer that secondary question. The safety counts are also descriptive, and the registry entries do not state whether reported deaths or adverse events were attributed to study medication.

Finally, the result should be read with the limits of registry reporting in mind. Details of missing-data handling, multiplicity, the full statistical plan, and the rationale for enrollment are not reported in the registry record. Those omissions limit how fully readers can assess the robustness and precision of the reported comparison.

14. Why This Trial Matters Statistically

This trial illustrates why the endpoint definition matters as much as the endpoint label. “Time to sustained clinical recovery” is operationalized through diary-based symptom change and temperature conditions. That specificity helps clarify what the analysis counts as recovery, while also showing that the result applies to this defined event rather than to every possible meaning of feeling better.

It also highlights the difference between a descriptive summary and an inferential comparison. The medians describe the observed groups, while the Gehan-Wilcoxon p-value summarizes a test of their time-to-event data. Neither alone provides the size and precision of a treatment effect. For that, an effect estimate and its uncertainty would be useful, but those are not listed for the primary comparison in the registry record.

The trial is also a reminder that a large enrollment entry does not automatically explain the analysis population for a particular outcome. The registry lists enrollment as 935 and separately defines the ITTI population and posts endpoint group counts. The record does not explain the relationship among these entries, so they should be kept distinct rather than treated as interchangeable.

Finally, safety counts, secondary outcomes, multiplicity, and missing-data procedures all contribute to interpretation. The registry lists serious and other adverse event counts, but the registry entries do not establish causality. It lists a secondary outcome without its result in the registry entries, and does not report several details needed to assess the analysis fully. Together, these features show why a p-value is only one part of a trial’s statistical account.

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 primary comparison is reported as a Gehan-Wilcoxon test, but no hazard ratio, confidence interval for the treatment contrast, or test statistic is listed in the registry record for this analysis. The posted medians and their accompanying values do not permit an independent reconstruction of that test. As a result, the magnitude and precision of the between-group effect cannot be directly characterized from these summaries.

The registry identifies the endpoint dispersion as an inter-quartile range and gives accompanying lower and upper values for each arm. The registry entries do not clarify whether those entries are the quartiles themselves or provide another kind of interval. They should not be interpreted as a confidence interval for the treatment contrast. The timeframe is up to 21 days, and no participant-level event times are available in the registry entries.

The primary efficacy population is defined by a positive baseline RT-PCR result. The registry entries do not explain the relationship between that population and the enrollment entry, or why the posted endpoint counts are as listed. Missing-data methods, censoring procedures, protocol deviations, and sensitivity analyses are not reported in the registry record. This limits assessment of how incomplete or unavailable endpoint information may have influenced the summaries.

The record lists 11 outcome measures and 13 analyses but does not report a multiplicity strategy, analysis hierarchy, or interim-analysis plan. The statistical analysis plan and sample-size justification are also not reported in the registry record. The secondary outcome of progression to severe COVID-19 is named, but its result is not included in the registry record for this analysis.

Safety information is reported as counts and selected serious adverse event terms. The registry entries do not state event attribution, the detailed definitions used for each category, or comparative statistical analyses. These limitations do not invalidate the posted entries, but they constrain the conclusions that can be drawn from them.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Trial to Evaluate Efficacy and Safety of Nitazoxanide in the Treatment of Mild or Moderate COVID-19 — NCT04486313.
• Registry record last updated 2024-03-29; status last verified 2024-03.
• Results first posted 2024-03-29.
• 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 record describes a completed Phase 3, randomized, double-blind, placebo-controlled, parallel-group trial sponsored by Romark Laboratories L.C. It lists COVID-19 as the condition, enrollment of 935, a start date of 2020-08-13, and a primary completion date of 2021-02-08.

The primary endpoint was time to sustained clinical recovery, measured in days for up to 21 days, in the ITTI population of subjects positive for SARS-CoV-2 by RT-PCR at baseline. The listed medians are 13.3 days for nitazoxanide and 12.4 days for placebo. The sponsor-reported Gehan-Wilcoxon test has p = 0.879.

The registry lists progression to severe COVID-19 as a secondary outcome and reports safety counts for serious and other adverse events. A recalculated primary comparison is not estimable from the posted summary statistics. Important details, including missing-data handling, multiplicity procedures, and the full statistical plan, are not reported in the registry record.