Efficacy and Safety of Molnupiravir (MK-4482) in Non-Hospitalized Adult Participants With COVID-19 (MK-4482-002): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Efficacy and Safety of Molnupiravir (MK-4482) in Non-Hospitalized Adult Participants With COVID-19 (MK-4482-002) (NCT04575597), a randomized phase 2, phase 3 study. The primary outcome is Percentage of Participants Who Were Hospitalized and/or Died Through Day 29 (Primary Pre-specified Analysis).

2. Clinical Question

The primary endpoint combines all-cause hospitalization and death from any cause through Day 29. The registry defines hospitalization as at least 24 hours of acute care in a hospital or similar acute care facility. Participants whose survival status was unknown at Day 29 were treated as having an event for this endpoint.

The trial record describes a superiority hypothesis for molnupiravir versus placebo on this outcome. It was a randomized, double-blind, placebo-controlled, parallel-group study with a phase 2/3 designation. Part 1 included three molnupiravir dose groups and a placebo group; Part 2 included an 800 mg molnupiravir group and a placebo group. The registry record does not report the original sample-size justification or the full statistical plan.

3. Trial Design

The registry record describes NCT04575597 as a randomized, double-blind, placebo-controlled, parallel-group phase 2/3 study sponsored by Merck Sharp & Dohme LLC. It was listed as completed, with enrollment of 1735. The recorded study start date is 2020-10-19 and the primary completion date is 2022-05-05.

The study had two parts. Part 1 compared molnupiravir at 200 mg, 400 mg, and 800 mg with matching placebo. Part 2 compared 800 mg molnupiravir with matching placebo. Each regimen was administered orally every 12 hours for 5 days, for 10 doses total. The registry describes the Part 2 dose as the dose selected for that part.

For the primary outcome, the Part 1 analysis population was randomized participants who received at least one dose and were not hospitalized before the first dose. The registry description also refers to Part 2 participants who had reached Day 29 by the prespecified futility/early efficacy analysis. The record does not report further detail about how participants were allocated between the two parts or the operational reason for the different analysis populations.

Alongside the primary endpoint, the registry lists symptom-resolution outcomes, including time to sustained resolution or improvement of targeted symptoms, and safety outcomes. The primary endpoint was the percentage hospitalized and/or dead through Day 29. The registry record does not report the original design assumptions, the planned statistical model for every endpoint, or a full account of interim decision procedures. Those details should not be inferred from the phase designation or the observed results.

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: All randomized participants in Part 1 who received at least one dose of study intervention and were not hospitalized prior to the administration of the first dose of study intervention, and all randomized participants in Part 2 who had reached Day 29 by the pre-specified futility/early efficacy analysis who received at least one dose of study intervention and were not hospitalized prior to the administration of the first dose of study intervention, were analyzed.

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.

Only the first 6 of 7 groups are shown.

6. Primary Endpoint

The registry lists the primary outcome as Percentage of Participants Who Were Hospitalized and/or Died Through Day 29 (Primary Pre-specified Analysis), measured over Up to 29 days.

The primary endpoint is the percentage of participants hospitalized and/or dead through Day 29. It combines all-cause hospitalization with death due to any cause, so a participant meeting either condition is counted as having the event. Hospitalization required at least 24 hours of acute care in a hospital or similar acute care facility. Unknown survival status at Day 29 was treated as an event in the endpoint definition.

This kind of binary endpoint records whether an event occurred within a specified period rather than the exact time at which it occurred. It can express the frequency of a clinically important event in each group and permit a comparison of event rates. The combined endpoint captures more events than either component alone might capture, but it also means that the overall result does not by itself show whether any group difference arose from hospitalization, death, or both. Component-specific primary results are not reported in the registry record.

The registry states that the primary hypothesis was superiority of molnupiravir to placebo. The result therefore needs to be read in light of the sponsor’s stated comparison, the endpoint definition, and the analysis population. The posted arm percentages describe event frequency; uncertainty about the comparison is best assessed using the sponsor-reported estimate and confidence interval, rather than treating the percentages alone as a definitive comparison.

7. Results Posted in the Registry

• Part 1: Molnupiravir 200 mg: 1.4% (1/74) versus 5.4% (4/74) on control; risk difference -4.1 percentage points (95% CI -9.8 to 1.7), risk ratio 0.25 (0.03–2.18); unadjusted two-proportion z-test p = 0.172.
• Part 1: Molnupiravir 400 mg: 3.9% (3/77) versus 5.4% (4/74) on control; risk difference -1.5 percentage points (95% CI -8.2 to 5.2), risk ratio 0.72 (0.17–3.11); unadjusted two-proportion z-test p = 0.659.
• Part 1: Molnupiravir 800 mg: 4.1% (3/74) versus 5.4% (4/74) on control; risk difference -1.4 percentage points (95% CI -8.2 to 5.5), risk ratio 0.75 (0.17–3.24); unadjusted two-proportion z-test p = 0.699.
• Part 2: Molnupiravir 800 mg: 7.3% (28/385) versus 5.4% (4/74) on control; risk difference 1.9 percentage points (95% CI -3.9 to 7.6), risk ratio 1.35 (0.49–3.72); unadjusted two-proportion z-test p = 0.563.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Time to Sustained Resolution or Improvement of Each Targeted COVID-19 Sign/Symptom - Cough

• Endpoint type: Time to event: compared as a hazard ratio.
• Test: Hazard ratio as reported by the sponsor; it was not recalculated.
• Result:
  • Part 1: Molnupiravir 200 mg vs Part 1: Placebo: hazard ratio 0.74 (95% CI 0.48–1.13) (as reported in the registry).
  • Part 1: Molnupiravir 400 mg vs Part 1: Placebo: hazard ratio 0.70 (95% CI 0.46–1.08) (as reported in the registry).
• Interpretation:
  • The 95% confidence interval includes no difference (a ratio of 1), so these data do not show a statistically significant difference between Part 1: Molnupiravir 200 mg vs Part 1: Placebo and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference (a ratio of 1), so these data do not show a statistically significant difference between Part 1: Molnupiravir 400 mg vs Part 1: Placebo and the control. The interval gives the range of effects that remain compatible with the data.

Time to Sustained Resolution or Improvement of Each Targeted COVID-19 Sign/Symptom - Sore Throat

• Endpoint type: Time to event: compared as a hazard ratio.
• Test: Hazard ratio as reported by the sponsor; it was not recalculated.
• Result:
  • Part 1: Molnupiravir 200 mg vs Part 1: Placebo: hazard ratio 1.49 (95% CI 0.80–2.76) (as reported in the registry).
  • Part 1: Molnupiravir 400 mg vs Part 1: Placebo: hazard ratio 1.20 (95% CI 0.69–2.09) (as reported in the registry).
• Interpretation:
  • The 95% confidence interval includes no difference (a ratio of 1), so these data do not show a statistically significant difference between Part 1: Molnupiravir 200 mg vs Part 1: Placebo and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference (a ratio of 1), so these data do not show a statistically significant difference between Part 1: Molnupiravir 400 mg vs Part 1: Placebo and the control. The interval gives the range of effects that remain compatible with the data.

Time to Sustained Resolution or Improvement of Each Targeted COVID-19 Sign/Symptom - Nasal Congestion

• Endpoint type: Time to event: compared as a hazard ratio.
• Test: Hazard ratio as reported by the sponsor; it was not recalculated.
• Result:
  • Part 1: Molnupiravir 200 mg vs Part 1: Placebo: hazard ratio 0.95 (95% CI 0.62–1.46) (as reported in the registry).
  • Part 1: Molnupiravir 400 mg vs Part 1: Placebo: hazard ratio 1.16 (95% CI 0.75–1.79) (as reported in the registry).
• Interpretation:
  • The 95% confidence interval includes no difference (a ratio of 1), so these data do not show a statistically significant difference between Part 1: Molnupiravir 200 mg vs Part 1: Placebo and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference (a ratio of 1), so these data do not show a statistically significant difference between Part 1: Molnupiravir 400 mg vs Part 1: Placebo and the control. The interval gives the range of effects that remain compatible with the data.

The registry reports symptom-resolution outcomes for multiple targeted COVID-19 symptoms, with results posted for 35 secondary outcomes. The available results include median times and confidence intervals, along with selected reported hazard-ratio comparisons. Across the secondary outcomes, the estimates do not present a single uniform pattern that can be summarized as a consistent shift in symptom-resolution timing. The uncertainty around many comparisons also limits how confidently differences can be distinguished from no difference.

These outcomes complement, but do not replace, the primary endpoint. The primary endpoint concerns hospitalization and/or death through Day 29, whereas symptom outcomes concern the time to sustained resolution or improvement. They address different questions and have different statistical scales. A pattern in symptoms cannot confirm or refute the primary result, and the primary result does not establish a particular effect on each individual symptom.

Secondary findings generally carry less weight when many outcomes and comparisons are examined, especially if results are unadjusted. Multiple testing increases the chance of observing an apparently notable result by chance. The registry record does not report whether or how multiplicity was controlled for the secondary outcomes. The reported secondary estimates should therefore be treated as supportive or exploratory unless the registry documents a prespecified testing strategy. No broad conclusion about all symptoms should be drawn from selected comparisons alone.

The registry lists eight targeted symptoms, including cough, sore throat, nasal congestion, rhinorrhea, shortness of breath or difficulty breathing, muscle or body aches, fatigue, and feeling hot or feverish. Results for each outcome should be read with its own sample and interval, rather than inferred from another symptom or from the primary endpoint.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• Time to Sustained Resolution or Improvement of Each Targeted COVID-19 Sign/Symptom - Cough
• Time to Sustained Resolution or Improvement of Each Targeted COVID-19 Sign/Symptom - Sore Throat
• Time to Sustained Resolution or Improvement of Each Targeted COVID-19 Sign/Symptom - Nasal Congestion
• Time to Sustained Resolution or Improvement of Each Targeted COVID-19 Sign/Symptom - Rhinorrhea
• Time to Sustained Resolution or Improvement of Each Targeted COVID-19 Sign/Symptom - Shortness of Breath or Difficulty Breathing
• Time to Sustained Resolution or Improvement of Each Targeted COVID-19 Sign/Symptom - Muscles or Body Aches
• Time to Sustained Resolution or Improvement of Each Targeted COVID-19 Sign/Symptom - Fatigue
• Time to Sustained Resolution or Improvement of Each Targeted COVID-19 Sign/Symptom - Feeling Hot or Feverish

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is binary: each participant either met the endpoint through Day 29 or did not. For each arm, the posted percentage is the proportion classified as having the event. A risk difference expresses the first group’s event rate minus the comparator group’s rate. Negative values indicate a lower event rate in the first group under that subtraction order; positive values indicate a higher event rate. The scale is percentage points when rates are reported as percentages.

What does a confidence interval describe?

A confidence interval gives a range of estimates that are reasonably compatible with the observed data and the method used to calculate the interval. It represents uncertainty in the estimated effect, not a range containing a specified proportion of individual participant outcomes. A narrower interval indicates greater precision on that scale; a wider interval indicates greater uncertainty. Whether an interval crosses zero is informative for a difference: zero corresponds to no difference on that scale. This is a general statistical interpretation and does not establish whether an effect is important in practice.

Why distinguish the sponsor analysis from a simple recalculation?

The posted recalculations are unadjusted calculations based on summary arm statistics. They can help readers see how a simple comparison of observed rates behaves, but they need not reproduce an analysis that uses the trial’s prespecified population, method, or comparator. The registry reports a Miettinen & Nurminen method for the Part 2 sponsor comparison. The method used for the other sponsor-reported rate differences is not stated. Since the discrepancy flag is true, the sponsor’s own analysis takes precedence for the trial’s reported primary comparison.

This distinction is especially important because the recalculated Part 2 comparisons in the posted data use Part 1 placebo as the reference, while the sponsor-reported Part 2 superiority analysis compares the Part 2 molnupiravir and placebo groups. A comparison involving a different control group answers a different question. It is not a substitute for the randomized Part 2 comparison.

How should the p-value be read?

A p-value assesses how incompatible the observed data are with a specified null hypothesis, assuming the statistical model and analysis conditions apply. It is not the probability that the null hypothesis is true, nor does it describe the probability that the result will be replicated. The sponsor-reported p-value for the Part 2 primary comparison accompanies its reported superiority analysis. The recalculated unadjusted p-values are not replacements for that sponsor analysis, particularly given the comparator discrepancy.

The registry also reports time-to-symptom-resolution outcomes. A time-to-event analysis considers when an event occurs and can account for participants whose event time is not observed within follow-up, depending on the analysis method. Hazard ratios compare event rates over time under a model; their interpretation depends on the event definition and analysis assumptions. A hazard ratio is not a difference in median days. The symptom results include reported hazard ratios for selected Part 1 comparisons, but the registry record does not identify the methods used to calculate them.

Why can over-dispersion matter?

General statistical context: for binary outcomes, a model may assume a particular relationship between the mean event probability and the variability of observed outcomes. Over-dispersion means the observed variability is greater than the model assumes. If not handled appropriately, uncertainty estimates can be too small and intervals too narrow. The registry record does not report whether over-dispersion was assessed or relevant to the primary analysis; it should not be assumed to have occurred in this study.

Likewise, randomization supports comparison of assigned groups, but the analysis population and handling of missing information still matter. The primary endpoint’s description specifies that unknown survival status at Day 29 was classified as an event. Other missing-data procedures, covariate adjustments, and sensitivity analyses are not reported in the registry record. These details cannot be reconstructed from the arm percentages or the comparison estimates alone.

The registry includes many outcome measures and analyses. When numerous outcomes or comparisons are examined, the chance of at least one apparently unusual result can increase if each is assessed without adjustment. Whether the sponsor controlled this multiplicity, and by what method, is not reported in the registry record. Consequently, the primary sponsor-reported analysis has a different evidential status from exploratory or unadjusted secondary comparisons.

11. Confidence Intervals and Precision

The Part 1 sponsor-reported rate differences have 95% confidence intervals that include zero. For the 200 mg comparison, the reported difference is -4.1 percentage points, with an interval from -12.2 to 2.5. For 400 mg, it is -1.5, with an interval from -9.9 to 6.2; for 800 mg, it is -1.3, with an interval from -9.6 to 6.4. These intervals allow for a range of possible differences, including no difference, and do not pin down the direction of the Part 1 effect precisely.

For the sponsor-reported Part 2 comparison, the difference is -6.8 percentage points and the 95% confidence interval ranges from -11.3 to -2.4. This interval excludes zero, but still spans multiple possible magnitudes of difference. Precision is not the same as importance: the interval describes statistical uncertainty, while the registry record does not specify a threshold for a clinically important difference.

The width and position of a confidence interval depend on the number of observed events, group sizes, and the method used. With a binary endpoint, the amount of information is influenced by event frequency as well as the number of participants. Sparse event counts can leave estimates uncertain even when the overall enrollment is substantial. The Part 1 and Part 2 populations and comparisons should be considered separately rather than using overall enrollment as a proxy for precision in any one contrast.

For secondary time-to-event results, the reported 95% intervals around hazard ratios are also broad in some comparisons and include the value representing no relative difference. The registry does not report p-values for those hazard-ratio estimates. Thus, the point estimates alone do not establish a reliable difference in time to symptom resolution.

12. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The sample-size figures are planning illustrations based on the observed effect, not the trial’s original design assumptions. For binary outcomes, the figures are participants per arm. For time-to-event outcomes, the figures are total events. The planning assumptions are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; each scenario targets the stated fraction of the observed effect. The trial’s own original sample-size assumptions are not reported in the registry record.

The 80% power column describes the planned number of participants per arm, or total events for time-to-event outcomes, under which a study would have an 80% chance of detecting the specified effect if that effect were present and the planning assumptions held. The 90% power column uses a higher chance of detection under those same assumptions, and therefore generally requires more information. These are design-planning quantities, not a description of how likely the observed result is to be true.

Rows that plan for 75% or 50% of the observed effect use progressively smaller effects as the target. Smaller effects are harder to distinguish from random variation and therefore require more participants or events. The figures depend on treating the observed effect as a planning reference; an observed effect can itself be imprecise, especially when event counts are limited. They should be read as illustrations of how sample-size requirements change with the target effect, not as evidence that the trial had adequate or inadequate power for its original purpose.

This is not observed power. The registry record does not report the assumptions behind the original trial design, so these illustrative figures cannot establish what power the sponsor planned or what power the completed trial achieved.

13. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, p₀ and p₁ are the control and treatment proportions, and p̄ is their average.

Worked calculation for this trial

1. Control proportion p₀ = 0.054; treatment proportion p₁ = 0.014; average p̄ = 0.034.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.2555 = 0.5008.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.2539 = 0.2137.
4. n = (0.5008 + 0.2137)² / (-0.0405)² = 310.6, rounded up to 311 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
p0 <- 0.0540540540540541                          # control proportion
p1 <- 0.0135135135135135                          # treatment proportion
n_per_arm <- function(power, frac = 1) {
  q1 <- p0 + (p1 - p0) * frac               # smaller effect = fraction of the observed difference
  pbar <- (p0 + q1) / 2
  z_b <- qnorm(power)
  ceiling(round((z_a * sqrt(2 * pbar * (1 - pbar)) + z_b * sqrt(p0 * (1 - p0) + q1 * (1 - q1)))^2 / (q1 - p0)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (unrounded n per arm)
cat("power.prop.test:", round(power.prop.test(p1 = p0, p2 = p1, power = 0.80)$n, 1), "\n")

# Simulation: how often does the z-test reject at the 80%-power sample size?
set.seed(4575597)
n    <- n_per_arm(0.80)
reps <- 20000
x0 <- rbinom(reps, n, p0); x1 <- rbinom(reps, n, p1)
ph <- (x0 + x1) / (2 * n)
z  <- ((x1 - x0) / n) / sqrt(ph * (1 - ph) * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(z) > z_a, na.rm = TRUE), reps))

Output from running the code

100% of observed effect: n per arm = 311 (80% power), 416 (90% power)
 75% of observed effect: n per arm = 633 (80% power), 847 (90% power)
 50% of observed effect: n per arm = 1,604 (80% power), 2,146 (90% power)
power.prop.test: 310.6 
Simulated power at n = 311 per arm: 0.830 (20000 replicates)

The simulation draws many pairs of event counts at the planned sample size and counts how often the two-proportion z-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

14. Analyses Reported by the Sponsor

The sponsor’s Part 2 primary analysis compares the Part 2 molnupiravir 800 mg arm directly with the Part 2 placebo arm. It reports a difference in rates of -6.8 percentage points, a 95% confidence interval from -11.3 to -2.4, and p = 0.001 using the Miettinen & Nurminen method. The registry identifies this as a superiority analysis. This is the relevant sponsor-reported comparison for the Part 2 primary outcome.

The simple recalculations do not reproduce that contrast: their reference group is Part 1 placebo. They are unadjusted recalculations from posted summary statistics and therefore should not be treated as the sponsor’s primary analysis. The difference in comparator is material, and the record marks a discrepancy. Under that condition, the sponsor’s own analysis takes precedence.

For the Part 1 comparisons, the sponsor reports rate differences and 95% confidence intervals but no p-values in the registry record. The recalculated unadjusted comparisons are not a substitute for any unreported sponsor analysis method or multiplicity procedure. Readers should keep the sponsor-reported trial analysis distinct from the supplementary recalculations and avoid using the latter to make a stronger claim than the posted sponsor result supports.

15. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when a study assesses multiple outcomes, treatment groups, time points, or statistical comparisons. If each test is interpreted separately without accounting for the full set, the chance of at least one result appearing statistically notable by chance can increase. This is a general statistical concern, not evidence that any particular result in this trial is false.

The registry record lists 39 outcome measures and 142 analyses, as well as numerous secondary outcomes. It identifies the primary outcome as a prespecified analysis and reports a sponsor superiority analysis for the Part 2 primary comparison. However, whether a formal multiplicity strategy controlled the overall error rate across dose groups, study parts, secondary outcomes, and analyses is not reported in the registry record. The record also does not report the full sequence of hypothesis tests or whether testing of secondary outcomes was conditional on the primary result.

The endpoint description states that unknown survival status at Day 29 was treated as failure for the primary outcome. That is a specific rule for the primary endpoint. The registry record does not report other details of missing-data handling, such as procedures for missing symptom assessments or sensitivity analyses for alternative assumptions about missing outcomes.

The primary analysis population is described differently for Part 1 and Part 2. Part 1 includes randomized participants who received at least one dose and were not hospitalized before the first dose. Part 2 includes randomized participants who had reached Day 29 by the prespecified futility/early efficacy analysis. The full analysis-population specification and the handling of participants outside those criteria are not reported in the registry record.

These omissions limit the ability to independently assess multiplicity, missing-data sensitivity, and the full analysis pathway. They do not negate the posted results, but they matter when judging how much confirmatory weight to place on multiple comparisons beyond the sponsor-reported primary analysis.

16. 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 5% 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 (7.6%), covid-19 pneumonia (6.1%), pneumonia (1.4%), diabetic metabolic decompensation (1.4%), pulmonary embolism (1.4%).

The registry reports serious adverse events and other adverse events separately by study part and arm. In Part 1, serious adverse events were reported for 1 participant in the 200 mg group, 3 in the 400 mg group, 4 in the 800 mg group, and 5 in the placebo group. The corresponding Part 1 groups each report their own at-risk count. In Part 2, serious adverse events were reported for 51 participants receiving molnupiravir 800 mg and 67 receiving placebo; the at-risk counts were 710 and 701, respectively. Deaths are reported separately in the safety data.

These are descriptive safety counts, not an adjusted comparison of risk. The safety at-risk counts differ from the primary-outcome arm sizes, so they should not be interchanged. The registry lists COVID-19 and COVID-19 pneumonia among the most frequently reported serious-event terms, with additional listed terms including pneumonia, diabetic metabolic decompensation, pulmonary embolism, and peripheral vascular disorder. The registry record does not provide a causal assessment for each event in the information. Counts alone do not establish that an event was caused by study treatment or that one group had a statistically reliable safety difference.

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

The posted results show that the sponsor-reported Part 2 primary comparison favored the molnupiravir group in the direction of a lower percentage hospitalized and/or dead through Day 29. Its confidence interval excludes zero, and the registry reports a superiority analysis. This describes the trial’s reported result for the specified population, endpoint, and comparison; it is not a general claim about every participant or every clinical setting.

The Part 1 sponsor-reported comparisons have confidence intervals that include zero, leaving substantial uncertainty about the size and direction of those differences. Part 1 and Part 2 should not be pooled or treated as interchangeable based solely on the listed arm percentages. The study parts have separate group structures, and the sponsor’s direct Part 2 comparison is the relevant contrast for that part.

The endpoint combines hospitalization and death, so the overall finding does not reveal which component contributed to the observed difference. It also does not address outcomes after Day 29. Symptom-resolution endpoints ask separate questions, and the registry’s secondary results do not replace the primary analysis. Safety counts are descriptive and do not, on their own, establish a causal relationship between treatment and an event.

Statistical significance does not settle clinical importance, applicability to populations outside the recorded study, or the balance of benefits and harms. The registry record does not report all details needed to assess those issues, including the full analysis plan, multiplicity handling, and all missing-data procedures. The findings should therefore be understood as the results of this randomized study as reported in the registry, with uncertainty and limitations considered alongside the point estimates.

18. Why This Trial Matters Statistically

This trial illustrates why the comparison underlying an estimate matters as much as the number itself. A simple rate difference can be calculated from posted arm summaries, but the chosen reference group determines the question being answered. In this record, the Part 2 sponsor analysis compares the Part 2 treatment and placebo groups, while the supplementary recalculations use Part 1 placebo as their reference. Treating these as the same contrast would misrepresent the evidence.

It also shows how confidence intervals help distinguish a numerical difference from a precise conclusion. The Part 1 point estimates are mostly on one side of zero, yet their intervals include zero. The sponsor’s Part 2 interval excludes zero, while still expressing a range of possible effect magnitudes. Neither the direction of a point estimate nor a p-value should be read without the interval and the analysis context.

Finally, this record demonstrates the importance of separating primary, secondary, and safety evidence. A prespecified primary endpoint has a defined role in the trial question; numerous secondary outcomes raise multiplicity concerns; and descriptive adverse-event counts do not establish treatment causation. Together, these distinctions help clinical researchers interpret registry results without confusing a reported association, its precision, and its broader implications.

19. 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.
• Participant counts were reconstructed from posted percentages and arm sizes, so small rounding differences are possible.

The registry record does not report the original sample-size justification or the trial’s full statistical plan. The planning illustrations based on observed effects cannot reconstruct the original design assumptions. The record also does not explain the full statistical approach used for every reported comparison, and methods for the Part 1 sponsor-reported rate differences are not stated.

The primary endpoint combines all-cause hospitalization and death. Component-specific results are not reported in the registry record, so the contribution of each component cannot be separated. Unknown survival status at Day 29 was treated as an event, but other missing-data procedures and sensitivity analyses are not reported.

The primary analysis populations are described differently for Part 1 and Part 2. The registry’s Part 2 description references participants who had reached Day 29 by the prespecified futility/early efficacy analysis, but does not report the complete operational details of that analysis population or its timing. Although the record mentions a futility/early efficacy analysis, the broader interim-analysis procedures are not reported.

The discrepancy between the simple recalculations and the sponsor’s Part 2 analysis is a further interpretive limitation. The recalculated comparisons use Part 1 placebo, while the sponsor’s primary Part 2 comparison uses Part 2 placebo. The sponsor’s own analysis takes precedence; the recalculations should not be used as an alternative estimate of the same comparison. Finally, the registry does not report whether multiplicity was controlled across the large set of outcomes and analyses, limiting interpretation of secondary findings as confirmatory evidence.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Efficacy and Safety of Molnupiravir (MK-4482) in Non-Hospitalized Adult Participants With COVID-19 (MK-4482-002) — NCT04575597.
• Registry record last updated 2023-06-28; status last verified 2023-05.
• Results first posted 2023-06-28.
• 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.

23. Record Summary

NCT04575597 is a completed randomized, double-blind, placebo-controlled, parallel-group phase 2/3 study of molnupiravir in non-hospitalized adults with COVID-19. The registry lists enrollment of 1735 and describes Part 1 dose groups of 200 mg, 400 mg, and 800 mg, followed by a Part 2 comparison of 800 mg with matching placebo. Treatment was administered orally every 12 hours for 5 days.

The primary outcome was hospitalization and/or death through Day 29. The sponsor-reported Part 2 comparison found a difference in rates of -6.8 percentage points, with a 95% confidence interval from -11.3 to -2.4. The registry also reports symptom-resolution outcomes and safety data. The sponsor’s analysis is the relevant Part 2 comparison because the simple recalculations use a different placebo group. Full details of the original design assumptions, multiplicity procedures, and several analysis methods are not reported in the registry record.