Clearing the Fog: Is Hydroxychloroquine Effective in Reducing COVID-19 Progression: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Clearing the Fog: Is Hydroxychloroquine Effective in Reducing COVID-19 Progression (NCT04491994), a randomized phase 3 study. The primary outcome is Number of Participants With Progression.

2. Clinical Question

The primary outcome concerns clinical or laboratory signs of worsening after treatment began. Progression was defined using fever lasting beyond the stated threshold, shortness of breath during a ten-step walk test, specified laboratory changes, or infiltrates on chest X-ray. The definition applied irrespective of PCR status. Viral clearance was listed as a secondary outcome, but its results are not reported in the registry record.

The posted outcome counts are 11 participants in the intervention group and 5 in the standard-of-care group. The groups had different reported sizes, so the counts alone are not an appropriate comparison. The risk estimates account for the group denominators: the posted risks are 0.032 for the intervention group and 0.033 for standard of care. The reported risk difference is -0.002, with a confidence interval spanning values on either side of no difference.

This is a question about the recorded trial endpoint and its statistical uncertainty, not a broad assessment of HCQ or COVID-19 treatment. The registry record does not provide enough information to assess all aspects of the enrolled population, conduct, or analysis.

3. Trial Design

The registry identifies this as a phase III randomized controlled trial with parallel groups and no masking. Its stated purpose was to evaluate whether hydroxychloroquine could reduce progression of COVID-19 and achieve viral clearance. The recorded status is completed. The registry gives a planned enrollment of 540 and lists a start date of 2020-04-10 and a primary completion date of 2020-05-31.

Participants were assigned to standard of care or an intervention group. The standard-of-care description specifies oral vitamin C, alfacalcidiol, zinc, and paracetamol as required. The intervention description specifies HCQ in addition to supportive treatment, with a loading schedule on day zero followed by treatment through day five. The registry does not provide details about how the random sequence was generated or how allocation was concealed.

The primary outcome was the number of participants with progression, assessed over five days after treatment began. The listed endpoint is a participant count, and the posted arm results give both the number of participants in each group and the number meeting the endpoint. The registry also lists viral clearance as a secondary outcome, but does not report its result in the posted outcome information.

The record lists 151 participants in the standard-of-care group and 349 in the intervention group. It separately reports enrollment of 540; the registry record does not explain the relationship between the enrollment figure and the arm counts. It also does not report a trial flow, exclusions from analysis, adherence, or the number of participants with missing primary outcome data. These omissions limit assessment of how the randomized groups were followed and how the reported endpoint counts were assembled.

No masking was recorded. Consequently, treatment assignment was not concealed from participants or trial personnel through blinding as described in the record. Whether outcome assessors were separately masked is not reported in the registry record. Other design and conduct details—including the protocol, analysis plan, and recruitment criteria—are also 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: 672 confirmed PCR positive cases were assessed for eligibility. 132 did not meet selection criteria and excluded.Later,15 withdrew consent and 5 became symptomatic. During follow up 13 patients were found to be deviated from advised therapy and 7 were lost to follow up yielding a final study population of 500.

Overall Study

Reasons for not completing this period:

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: sample size calculation was done and 540 patients enrolled

6. Primary Endpoint

The registry lists the primary outcome as Number of Participants With Progression, measured over 5 days.

The primary endpoint is the number of participants with progression during the five-day period after treatment started. It is a binary participant-level outcome: each participant either met the progression definition or did not. The registry definition includes fever above the stated threshold for the stated duration, shortness of breath on a ten-step walk test, specified laboratory changes, or infiltrates on chest X-ray during treatment. The definition applies irrespective of PCR status.

A composite definition can capture several ways a participant’s condition might worsen, rather than relying on a single sign. That can be useful when the clinical question concerns progression that may be expressed through different symptoms, laboratory findings, or imaging. At the same time, the components may differ in clinical meaning and in how they are measured. The record does not report separate results for the individual components, or details about adjudication and measurement procedures.

The posted results report 5 participants with progression among 151 in the standard-of-care group, and 11 among 349 in the intervention group. Because the groups have different denominators, comparing event counts alone would not describe the relative frequency of the endpoint. The risk estimates use each group’s reported denominator. The primary endpoint is distinct from viral clearance, which is listed as a secondary outcome but has no result reported in the registry record.

The five-day timeframe provides a defined window for counting the endpoint, but the record does not describe follow-up beyond that window for this outcome. The endpoint therefore addresses recorded progression during the specified period, not later outcomes or every possible consequence of COVID-19.

7. Results Posted in the Registry

• Intervention Group: 3.2% (11/349) versus 3.3% (5/151) on control; risk difference -0.2 percentage points (95% CI -3.6 to 3.2), risk ratio 0.95 (0.34–2.69); unadjusted two-proportion z-test p = 0.926.

8. Statistical Interpretation

9. Statistical Methods Explained

For a binary endpoint, each participant is classified according to whether the event occurred. Here, the event is meeting the registry definition of progression during the stated five-day timeframe. The posted counts are 5 events among 151 participants in standard of care and 11 among 349 participants in the intervention group. A group risk expresses the event count relative to that group’s denominator. The posted risks are 0.033 and 0.032, respectively.

What does the risk difference describe?

The risk difference is the intervention-group risk minus the standard-of-care risk. Its reported estimate is -0.002. A value of zero corresponds to equal risks; a negative value indicates a lower observed risk in the intervention group, while a positive value indicates a higher observed risk. This is an absolute contrast, so it describes the difference between the two group risks rather than a proportional change. The confidence interval is -0.035 to 0.032, covering zero and values on either side. That breadth matters: the point estimate alone does not capture how uncertain the comparison is.

How are relative measures different?

The risk ratio compares the intervention risk with the standard-of-care risk as a ratio, while the odds ratio compares the groups’ odds of the event. The posted risk ratio is 0.952, with interval 0.337 to 2.692; the posted odds ratio is 0.950, with interval 0.324 to 2.784. For either measure, the value corresponding to no relative difference is one. Both intervals include that value and span estimates on either side. These measures answer a different question from the risk difference, and their numerical scales should not be conflated.

What does the p-value say?

The posted unadjusted p-value is 0.926. In general, a p-value describes how compatible the observed data are with a specified null hypothesis under the assumptions of a statistical test. It is not the probability that the null hypothesis is true, and it is not the probability that a treatment works. A large p-value does not prove equality or rule out clinically meaningful differences. The registry record does not describe the test or the prespecified analysis plan, so the value should be interpreted cautiously as an unadjusted comparison.

Why does the confidence interval matter?

In general, a confidence interval represents uncertainty in an estimate under a statistical procedure and its assumptions. Its endpoints show a range of effect estimates that are compatible with the observed data under that procedure. Here, the interval for the risk difference crosses zero, and the relative-measure intervals include their no-difference value. This means the posted data do not identify a single direction with precision. The interval is not a guarantee that the true effect lies within its endpoints, and its interpretation depends on the method and data quality.

The arm counts differ, and the posted group denominators are used to form the risks. Comparing the raw event counts without their denominators would be misleading. The registry record lists 540 as enrollment and separately lists arm counts of 151 and 349; it does not explain how those figures relate. The record also does not report the analysis population or whether every randomized participant contributed to the primary endpoint calculation.

The posted comparison is an unadjusted recalculation from the summary statistics. An unadjusted comparison does not account for covariates or other factors that could be relevant to the outcome. No adjusted estimates, subgroup analyses, or detailed model specifications are reported in the registry record. The record lists no analyses and does not contain an account of a sponsor’s primary statistical analysis. Statistical uncertainty in the recalculated results therefore should not be mistaken for a complete trial analysis.

For binary outcomes, the precision of estimated risks and contrasts depends on the observed event information as well as group sizes. When event counts are limited, estimates can remain uncertain, particularly for relative measures. General statistical methods may use different approximations or models, and the registry record does not specify which method generated each posted comparison. The displayed confidence intervals and p-value are the reported statistics; no additional method should be assumed.

In general, over-dispersion refers to variability greater than a model expects. It can make standard errors too small if ignored, producing intervals that are too narrow and tests that appear more decisive than warranted. Over-dispersion is relevant to some count-data models, but this endpoint is reported as participants with a binary event, and the registry record does not identify an over-dispersed model. It should not be treated as a feature of this trial’s analysis without documentation.

Randomization is intended, in general, to support a fair comparison by assigning participants to groups without basing assignment on their expected outcomes. The record identifies the trial as randomized but does not report the sequence-generation or concealment methods. No masking was recorded. The absence of masking can matter for outcomes that involve judgment or behavior, although the record does not provide enough detail to assess how endpoint ascertainment was conducted. These design considerations are separate from the arithmetic of the posted risks and contrasts.

10. Confidence Intervals and Precision

The most direct measure of precision for the absolute comparison is the risk-difference confidence interval. The estimate is -0.002, and its interval extends from -0.035 to 0.032. The interval crosses zero, so the posted data are compatible with no difference as well as with a difference in either direction. Its width indicates that the estimate is not highly precise. The point estimate should therefore not be read as a definitive indication of the direction or size of a treatment effect.

The relative measures also have broad intervals. The risk ratio is 0.952, with a confidence interval from 0.337 to 2.692. The odds ratio is 0.950, with a confidence interval from 0.324 to 2.784. Each interval includes the value representing no relative difference and spans both lower and higher values. The ratios’ point estimates being near that value does not remove the uncertainty expressed by their intervals.

Confidence intervals are general statistical tools for expressing sampling uncertainty, conditional on the method and assumptions used. They do not account automatically for missing data, measurement error, bias, or differences between the registry record and trial conduct. The record does not report the interval-generation method, missing-data handling, or analysis population. Precision in the displayed estimates cannot resolve those additional uncertainties.

The limited number of recorded progression events is relevant to the uncertainty of the comparison. The registry reports the event counts, but does not provide component-specific outcome counts or details about ascertainment. The interval is therefore the key context for interpreting the small difference between the posted risks. A result that is not statistically persuasive is not the same as evidence that the groups are identical; equally, a point estimate in one direction does not establish a reliable effect.

11. 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 planning figures are illustrations based on the observed effect, not a report of the trial’s original sample-size calculation. They assume a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. Each planning row uses the stated fraction of the observed effect as the effect to detect. The trial’s own original design assumptions and sample-size justification are not reported in the registry record.

The figures are participants per arm for 80% power and participants per arm for 90% power. In general, power is the probability that a statistical test will detect an effect of a specified size if that effect is truly present, under the planning assumptions. The 90% column reflects a higher target probability of detection than the 80% column, and therefore requires more participants per arm for the same assumed effect.

The rows describe planning for 100%, 75%, and 50% of the observed effect. For 100% of the observed effect, the figures are 193,209 participants per arm for 80% power and 258,652 for 90% power. For 75% of the observed effect, they are 345,531 and 462,567 participants per arm. For 50% of the observed effect, they are 782,048 and 1,046,939 participants per arm. The pattern illustrates a general principle: smaller effects are harder to distinguish from random variation and need more participants for the same target power.

These are planning illustrations, not observed power. They do not show the probability that the completed trial successfully detected an effect, and they are not a retrospective measure of the trial’s quality. The calculations depend on the stated assumptions and on using the observed effect as a reference. Because the observed risk difference is uncertain, the planning figures should be understood as hypothetical design scenarios rather than definitive requirements for a future study.

12. 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 = 0.033; treatment proportion p₁ = 0.032; average p̄ = 0.032.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.2501 = 0.4902.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.2501 = 0.2105.
4. n = (0.4902 + 0.2105)² / (-0.0016)² = 193,208.9, rounded up to 193,209 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 <- 0.033112582781457                          # control proportion
p1 <- 0.0315186246418338                          # 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(4491994)
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 = 193,209 (80% power), 258,652 (90% power)
 75% of observed effect: n per arm = 345,531 (80% power), 462,567 (90% power)
 50% of observed effect: n per arm = 782,048 (80% power), 1,046,939 (90% power)
power.prop.test: 193208.9 
Simulated power at n = 193,209 per arm: 0.793 (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.

13. Multiplicity, Missing Data and Analysis Populations

The registry lists the primary progression outcome and viral clearance as a secondary outcome. It does not report a result for viral clearance. The record does not describe whether there were additional planned outcomes, subgroup comparisons, interim analyses, or repeated looks at accumulating data. These details are not reported in the registry record.

Multiplicity is a general statistical issue that arises when several outcomes, comparisons, or analyses are tested. As the number of tests increases, the chance of finding at least one apparently unusual result by chance can increase unless the analysis plan accounts for the testing structure. The registry record does not describe a multiplicity adjustment or a hierarchy for interpreting outcomes. It is therefore not possible to determine from the record how multiple comparisons were handled.

The record lists no analyses and does not provide an analysis population definition. It does not state whether the primary comparison followed an intention-to-treat principle, whether participants were analyzed according to assignment, or whether any per-protocol or other population was used. These choices can affect interpretation, particularly when adherence, eligibility after randomization, or follow-up differ between groups.

Missing-data information is also absent. The registry does not report how many participants lacked a primary outcome assessment, whether missing outcomes were imputed, or whether sensitivity analyses were performed. A count-based endpoint can be affected if participants with unknown outcome status are treated differently from those with recorded outcomes. Without a participant flow or missingness description, the completeness of the posted endpoint counts cannot be assessed.

These omissions do not establish that the trial had missing data or multiplicity problems; they mean that the record does not report the information needed to evaluate them. No interim analysis plan, stopping rule, or detailed statistical analysis plan is reported in the registry record. The unadjusted comparison can be described, but these missing analysis details limit conclusions about how it relates to the trial’s intended inferential framework.

14. 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.

The registry reports serious adverse events for the listed groups. For the control group, 151 participants were listed as at risk and 0 serious adverse events were reported. For the intervention group, 349 participants were listed as at risk and 0 serious adverse events were reported. The record also lists 0 other adverse events and 0 deaths in each group.

These figures describe events reported in the registry record and should not be taken to mean that no adverse effects occurred. The record does not provide a detailed adverse-event collection method, event terms, severity grading beyond the serious-event category, follow-up duration for safety, or information about whether all participants were assessed. The absence of listed events is not proof that an intervention is free of risk.

The serious adverse event figures are separate from the primary progression endpoint. The registry’s progression definition includes clinical, laboratory, and imaging findings, while the adverse-event entries use adverse-event categories. They should not be combined or treated as equivalent outcomes. The posted safety information is limited to the recorded counts and does not support a comprehensive comparison of safety profiles.

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

The posted results show 5 participants with progression among 151 in the standard-of-care group and 11 among 349 in the intervention group during the five-day outcome window. The corresponding risks are 0.033 and 0.032, and the reported risk difference is -0.002. The point estimate is small and favors a lower recorded risk in the intervention group, but the confidence interval spans both negative and positive values. The data therefore do not precisely establish the direction or size of a difference.

The unadjusted p-value of 0.926 is not evidence that the groups are identical, and it is not the probability that treatment has no effect. The confidence intervals provide important context: they include the no-difference value and allow for estimates in either direction. The relative measures are similarly uncertain. The appropriate conclusion is about the limited precision of the posted comparison, not a categorical claim of benefit or lack of benefit.

The outcome is specific to the registry’s definition of progression and its five-day timeframe. It does not answer whether viral clearance differed, because that secondary result is not reported in the registry record. Nor does it describe outcomes beyond the stated window or all possible measures of COVID-19 progression. The record does not give enough information to evaluate missing data, outcome adjudication, or the trial’s complete analysis plan.

Randomization supports a comparative design in principle, but the record does not describe allocation procedures, masking beyond the notation that there was none, or participant flow. The reported enrollment and arm counts are listed separately, with no explanation of their relationship. Those gaps limit assessment of how closely the posted summaries reflect the intended randomized comparison.

These statistics should be read as a report of what was recorded and how uncertain the unadjusted comparison is. They do not provide medical advice, establish a mechanism, or prove that treatment does or does not work. A fuller interpretation would require trial documentation and outcome information not reported in the registry record.

16. Why This Trial Matters Statistically

This trial illustrates why event counts need their denominators. The intervention group has 11 recorded progression events and the standard-of-care group has 5, but the groups also have different reported sizes. The risks and risk difference provide a more informative comparison than the raw counts alone. Even with those denominators, a point estimate is not enough: the confidence interval shows the range of uncertainty around the estimated contrast.

The risk-difference interval crosses zero and extends to values on both sides. This makes the result a useful example of the distinction between a small observed difference and a precise estimate. A p-value of 0.926 does not prove no effect, just as the negative point estimate does not prove a reduction. Statistical interpretation depends on looking at the estimate, its uncertainty, the endpoint definition, and the design together.

The planning illustrations also show how assumed effect size affects study size. Smaller effects require more participants per arm to reach the same target power under the stated assumptions. These calculations are not the trial’s original design rationale and do not tell readers what the completed study’s power was. Keeping that distinction clear prevents a planning quantity from being mistaken for a result.

Finally, the registry record contains important reporting gaps: no analyses are listed, the secondary outcome result is absent, and details about missing data, allocation, and analysis populations are not reported. Recognizing these limits is part of statistical interpretation. A numerical comparison can be described carefully without treating the registry entry as a complete account of trial conduct or as a basis for a broad clinical conclusion.

17. 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.

Several trial-specific limitations follow from the information recorded. The study was unmasked, and the registry does not say whether endpoint assessors were masked or how progression criteria were assessed. The primary endpoint combines symptoms, laboratory findings, and imaging; separate component results and adjudication procedures are not reported. It is therefore not possible to determine whether the components contributed similarly to the posted event counts.

The record reports planned enrollment of 540 and arm counts of 151 and 349, but does not explain their relationship. It does not provide a participant flow, eligibility criteria, baseline characteristics, or reasons for any exclusions. The analysis population and the completeness of primary outcome follow-up are not reported. These omissions make it difficult to assess comparability after randomization and whether the posted denominators represent all relevant participants.

The trial’s allocation methods, protocol, original sample-size assumptions, and detailed statistical analysis plan are not reported in the registry record. Nor are adjusted analyses, interim analyses, or multiplicity handling described. The unadjusted estimates can be read from the posted summaries, but they cannot stand in for a complete assessment of the intended analysis.

The primary endpoint covers five days, and the registry provides no longer-term result for progression. Viral clearance is listed as secondary, but its result is not reported. Safety reporting is limited to the adverse-event counts, without detailed event terms or collection procedures. These reporting gaps restrict interpretation of both efficacy-related endpoints and safety information.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Clearing the Fog: Is Hydroxychloroquine Effective in Reducing COVID-19 Progression — NCT04491994.
• Registry record last updated 2020-08-21; status last verified 2020-08.
• Results first posted 2020-08-21.
• 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.

21. Record Summary

The registry describes NCT04491994 as a completed, randomized, parallel-group phase III trial with no masking. The stated purpose was to assess hydroxychloroquine for COVID-19 progression and viral clearance. Planned enrollment was 540. The recorded study start was 2020-04-10, and primary completion was 2020-05-31.

The primary endpoint was participants with progression over five days. The posted counts were 5 of 151 in standard of care and 11 of 349 in the intervention group. The reported risk difference was -0.002, with a confidence interval from -0.035 to 0.032 and an unadjusted p-value of 0.926. The interval includes no difference and supports a cautious reading of the imprecise comparison.

Viral clearance is listed as a secondary outcome, but its result is not reported in the registry record. The record lists no analyses and does not report the trial’s detailed analysis plan, missing-data handling, or analysis population. Serious adverse events and other adverse events were recorded as 0 in each group, as were deaths; the available safety detail is limited.