FEnofibRate as a Metabolic INtervention for COVID-19: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for FEnofibRate as a Metabolic INtervention for COVID-19 (NCT04517396), a randomized phase 2 study. The primary outcome is Primary Hierarchical Composite Endpoint.

2. Clinical Question

The trial is titled FEnofibRate as a Metabolic INtervention for COVID-19 and is registered as NCT04517396. Its stated aim was to assess the clinical impact of fenofibrate on clinical outcomes in patients with COVID-19. The registry summary describes the rationale in terms of metabolic changes associated with COVID-19, but that rationale is not itself evidence of a clinical treatment effect.

The primary endpoint is a hierarchical composite rather than a single event or symptom. It is intended to order overall clinical outcomes by severity using multiple dimensions. The registry record reports group-level medians and inter-quartile ranges for that score, but does not report a primary between-group analysis. The clinical question is therefore broader than whether any one component, such as death or respiratory support, differed on its own.

3. Trial Design

This was a phase 2, randomized, quadruple-masked, parallel-group trial sponsored by the University of Pennsylvania. The registry lists an enrollment of 701 and a COVID-19 condition. The study began on 2020-08-18, had a primary completion date of 2022-03-30, and is marked completed.

The two reported arms were Fenofibrate + Usual Care and Placebo + Usual Care. The fenofibrate arm description specifies 145 mg of Tricor or a dose-equivalent preparation. The trial summary describes the intended course as 10 days and notes dose adjustment in chronic kidney disease. The placebo arm received a matching placebo with usual care. The record lists fenofibrate/fenofibric acid, placebo, and usual care among the interventions, but presents these two combined arms for the randomized comparison.

Randomization is intended to distribute prognostic factors across groups on average, while masking is intended to limit behavior or assessment influenced by knowing the assigned treatment. The registry identifies the trial as quadruple-masked but does not specify in the information reported who was masked. The parallel model means the listed arms are compared as groups rather than as sequential treatment periods within the same participant.

The posted primary outcome is a hierarchical composite assessed over 30 days. A secondary outcome describes days alive and free from specified levels of respiratory support or intensive care, up to 30 days. The registry also lists a seven-category ordinal scale and a secondary hierarchical composite endpoint. Further details of the allocation process, eligibility criteria, and statistical analysis plan are not reported in the registry record.

4. Participant Flow and Accounting

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

Overall Study

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

6. Primary Endpoint

The registry lists the primary outcome as Primary Hierarchical Composite Endpoint, measured over 30 days.

The primary endpoint is called the Primary Hierarchical Composite Endpoint. It is described as a nonparametric global rank score: participants were ranked from worst to best clinical outcomes on a scale of 1 to 701. The ranking prioritizes time to death, then days supported by invasive mechanical ventilation or extracorporeal membrane oxygenation, then an oxygen-related measure, followed by outpatient-specific measures.

For outpatients who were subsequently hospitalized, the hierarchy uses days out of hospital during the 30-day period after randomization. For outpatients who were not hospitalized during that observation period, it uses the modified Borg dyspnea scale. The endpoint therefore combines outcomes of differing kinds and applies an ordering rule to reflect their relative clinical severity. It is not a simple count of events or an average of one common measurement.

Hierarchical endpoints can capture a range of outcomes while prioritizing more severe events. This can be useful when a trial seeks to compare overall clinical experience rather than focus on a single component. However, the resulting score depends on the ranking rules, and a group-level score does not show which individual component accounts for a difference. Component-specific results and a prespecified analysis are needed to interpret those contributions. The registry reports the endpoint over 30 days and gives medians and inter-quartile ranges by arm, but does not report the primary comparative analysis.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Statistical methods help distinguish descriptive summaries from comparisons between groups. For this trial, the registry reports descriptive statistics for the primary endpoint and a secondary unadjusted comparison for the days-based outcome. It does not report a sponsor analysis for the primary endpoint. Accordingly, the primary medians and inter-quartile ranges should not be treated as though they were an inferential test.

What does a median describe?

A median is the midpoint of an ordered set of observations: half of the observed values lie at or below it and half lie at or above it, subject to ties. It is less sensitive than a mean to extreme values. The inter-quartile range gives the middle portion of the observed distribution, from the lower to the upper quartile. These summaries describe the distribution but do not quantify the uncertainty in a between-group contrast unless an appropriate comparative analysis is also reported.

For the primary ranked severity score, the record gives a median and inter-quartile range for each arm. The ranking itself orders outcomes from worst to best using the stated hierarchy. A median score is not a direct estimate of a difference in survival, respiratory support, or another individual component. Component-level interpretation would require results for those components and a clear analysis plan.

What does a mean difference describe?

A mean difference compares the average outcome in one group with the average in another. The posted secondary result gives the mean difference for placebo relative to fenofibrate as -0.5 days. Its sign depends on the order of subtraction, so the comparison direction matters when reading it. The group means are 28.8 and 28.3 days, respectively; the comparison is reported with a 95% confidence interval and an unadjusted p-value.

A confidence interval describes the range of estimates compatible with the observed data under the statistical model and its assumptions. It communicates precision: a narrower interval indicates greater precision than a wider interval, all else equal. An interval that includes zero does not establish that the true difference is zero. Nor does it show that the groups are equivalent. Equivalence requires a prespecified margin and an analysis designed to evaluate whether the plausible differences fall within that margin; these are not reported in the registry record.

Why does the primary endpoint need its own analysis?

A hierarchical composite is analyzed according to the way its outcome ranking is constructed. The endpoint gives precedence to some outcomes over others, so an analysis needs to respect that ordering. The record describes how the score was generated, but does not report a comparative estimate, confidence interval, or p-value for it. It is therefore not possible to infer the result of a formal primary analysis from the arm-specific medians alone.

Nonparametric methods are commonly used when an outcome is ranked or when distributional assumptions for a conventional mean-based model are unsuitable. In general, such methods compare the ordering of outcomes rather than requiring that the outcome follow a particular distribution. The exact method used for this primary endpoint is not reported in the registry record, so no specific test or model should be attributed to the trial.

What does the unadjusted label mean?

The secondary mean comparison is reported as unadjusted. In general, an unadjusted comparison does not account for other covariates or for a multiplicity procedure across several outcomes. Such a result can be informative as a descriptive or exploratory comparison, but its interpretation depends on the analysis plan, the number of outcomes examined, and whether the result was prespecified. Those details are not reported in the registry record.

The record lists multiple secondary outcomes and reports no analyses in its analysis listing. The posted mean comparison should therefore not be mistaken for evidence that all secondary endpoints were tested under a complete, prespecified inferential framework. The registry does not state whether the estimate was planned, whether adjustments were made elsewhere, or how missing data were handled for this outcome.

How should planning illustrations be read?

Sample-size calculations based on an observed effect are planning illustrations, not observed power. The posted illustration uses a two-sided significance level of 0.05, equal allocation between arms, and a normal approximation. It considers planning for 100%, 75%, and 50% of the observed effect, with participants per arm for 80% power and participants per arm for 90% power. These values describe hypothetical future planning under those assumptions; they do not explain the original trial’s design or demonstrate the probability that the completed trial detected an effect.

Observed-effect planning is particularly sensitive to uncertainty in the estimate used. If an estimated effect is imprecise, calculations anchored to it can give unstable planning targets. A smaller assumed effect generally requires more participants to detect under the same framework, but no new sample-size values should be inferred beyond those posted. The trial’s own sample-size justification and statistical plan are not reported in the registry record.

10. Multiplicity, Missing Data and Analysis Populations

The registry record lists multiple secondary outcomes and identifies a primary hierarchical composite endpoint. It does not report a multiplicity adjustment or a testing hierarchy for the secondary outcomes. In general, multiplicity matters when several outcomes or comparisons are tested: interpreting each result as though it were the only test can overstate the strength of evidence. Without a reported plan, the inferential status of the secondary comparison is unclear.

The record also does not report a primary comparative analysis, despite giving descriptive primary endpoint values. It lists no reported analyses. As a result, information needed to assess the primary hypothesis—including the analysis method, the treatment-effect estimate, and its uncertainty—is not reported in the registry record. The unadjusted secondary comparison should not be used to fill that gap.

Missing-data methods and analysis populations are not reported in the registry record. The record does not state how missing outcome values were handled, whether all randomized participants were included in each analysis, or whether any analysis population differed from the randomized groups. These details matter because incomplete outcome data can affect both estimates and their precision, and because exclusions after randomization can compromise the benefits of random assignment.

The registry record does not report whether interim analyses occurred, whether a sample-size justification was used for the original design, or what statistical analysis plan governed the trial. These omissions do not establish that such procedures did or did not occur; they mean the information is not reported in the registry record. The posted sample-size illustration is based on the observed secondary effect and stated planning assumptions, so it should not be interpreted as the trial’s original sample-size rationale.

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 1% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: worsening dyspnea or hypoxia (8.9%), death (6.3%), intensive care unit transfer (4.6%), invasive mechanical ventilation (4.6%), other (4.6%).

The registry reports serious adverse-event counts of 46 for Fenofibrate + Usual Care and 61 for Placebo + Usual Care. The corresponding groups at risk are reported as 351 and 350. Other adverse events are reported as 74 and 55, respectively. The record also reports 19 deaths in the fenofibrate group and 22 in the placebo group. These are posted counts; they do not establish the cause of an event or whether it was related to the assigned intervention.

The most frequent listed serious-event terms include worsening dyspnea or hypoxia at 8.9%, death at 6.3%, intensive care unit transfer at 4.6%, invasive mechanical ventilation at 4.6%, other at 4.6%, and acute kidney injury at 3.4%. The registry record does not provide a treatment-specific breakdown for these listed terms in the information. They should not be read as comparative risks for each arm.

Safety counts are descriptive. A fair comparison requires consistent definitions, observation periods, and appropriate analyses, as well as context about underlying illness. The posted counts alone do not show whether differences are statistically reliable or attributable to treatment. The record does not report a comparative safety analysis or additional detail needed to interpret event timing and causality.

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

The record documents a randomized, masked comparison of fenofibrate plus usual care with placebo plus usual care in a phase 2 trial of COVID-19. It reports descriptive results for a primary hierarchical severity score and a comparative result for one secondary days-based outcome. Those results can be described, but the missing primary comparative analysis limits what can be concluded about the trial’s principal question.

The primary medians and inter-quartile ranges are similar as reported, but descriptive summaries alone do not establish whether the groups differed under the primary endpoint’s ranking framework. No primary treatment-effect estimate, confidence interval, or p-value is reported in the registry record. The result therefore cannot be characterized as a statistically established benefit, lack of benefit, or equivalence on the primary outcome.

The secondary mean comparison has an unadjusted p-value of 0.232 and a confidence interval from -1.319 to 0.319. That interval conveys uncertainty in the estimated mean difference, and it includes values on both sides of zero. The result does not prove that the groups are identical, nor does it establish that either group had a clinically important advantage. No threshold for clinical importance is reported.

These findings also do not identify the effect of fenofibrate apart from usual care, because the randomized comparison is between two treatment strategies that both include usual care. The registry record does not provide results for every listed secondary endpoint, a multiplicity plan, or details of missing-data handling and analysis populations. Those limits constrain interpretation of both efficacy and safety findings.

In general, a registry entry is one source of trial information, and the completeness of its results determines what can be evaluated from that record. Here, the posted data support a careful account of reported group summaries and one unadjusted secondary comparison. They do not support a definitive conclusion about the primary endpoint or a broad claim about clinical effectiveness.

13. Why This Trial Matters Statistically

This trial illustrates why endpoint definitions and analysis reporting matter as much as the treatment-group summaries. Its primary outcome is a hierarchical composite that combines mortality, respiratory support, oxygen-related measures, and outpatient-specific outcomes. Such endpoints can represent several dimensions of clinical severity, but their interpretation depends on the ranking rules and on a suitable analysis that compares the randomized groups.

The registry record gives arm-specific medians and inter-quartile ranges for the primary score but does not report the primary comparative result. This gap shows why descriptive values cannot substitute for an estimate of the treatment contrast and its uncertainty. Without the primary analysis, readers cannot assess whether the ranking outcome provides evidence of a group difference under the trial’s main question.

The record also illustrates the different roles of primary and secondary outcomes. A secondary result can add context, but it does not automatically confirm or contradict a primary result—particularly when the primary analysis is absent and the secondary analyses are unadjusted or incompletely reported. Multiplicity, prespecification, missing data, and analysis populations all affect how confidently results can be interpreted.

Finally, planning calculations based on an observed effect are not a substitute for the trial’s design rationale. The posted sample-size illustration is useful as a hypothetical planning exercise under stated assumptions, but it does not reveal the original sample-size justification or show observed power. Clear reporting of the planned analysis, all outcome results, and the handling of incomplete data is essential for evaluating both precision and the strength of evidence.

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 principal limitation in the registry record is the absence of a reported comparative analysis for the primary hierarchical endpoint. The endpoint description is detailed, and group-level medians and inter-quartile ranges are posted, but no primary effect estimate or inferential result is provided. This leaves the principal trial question unresolved by the reported primary results.

Comparative results are not reported for all listed secondary outcomes. The available days-based comparison is unadjusted, and the record does not state how it fits into a prespecified testing strategy or whether multiplicity was addressed. No analysis plan, missing-data method, or analysis-population definition is reported, limiting assessment of potential bias and precision.

The primary endpoint combines outcome domains, but the registry record does not give component-specific comparative results. Consequently, a group-level ranked score cannot identify which outcome domain may have contributed to the observed summaries. The record also does not provide further eligibility details in the posted endpoint description, or specify who was masked despite describing the trial as quadruple-masked.

Safety information is presented as counts and listed event terms without a reported comparative safety analysis or sufficient detail to infer causality. These limitations concern what is reported in the registry record; they do not establish that additional analyses or documentation do not exist elsewhere.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: FEnofibRate as a Metabolic INtervention for COVID-19 — NCT04517396.
• Registry record last updated 2023-03-24; status last verified 2023-03.
• Results first posted 2023-02-27.
• 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

NCT04517396, titled FEnofibRate as a Metabolic INtervention for COVID-19, is a completed phase 2 randomized, quadruple-masked, parallel-group trial sponsored by the University of Pennsylvania. The registry reports enrollment of 701 and compares Fenofibrate + Usual Care with Placebo + Usual Care. The fenofibrate regimen is described as 145 mg of Tricor or a dose-equivalent preparation, with a 10-day course described in the trial summary and dose adjustment in chronic kidney disease.

The primary outcome is a 30-day hierarchical ranked severity score. The registry reports group medians and inter-quartile ranges but no primary comparative analysis. One secondary days-based outcome has an unadjusted mean comparison; results for other listed secondary outcomes are not reported in the registry record. The record also reports adverse-event counts, but not a comparative safety analysis. Important details about multiplicity, missing data, analysis populations, and the trial’s statistical plan are not reported in the registry record.