Hydroxychloroquine Post Exposure Prophylaxis for Coronavirus Disease (COVID-19): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Hydroxychloroquine Post Exposure Prophylaxis for Coronavirus Disease (COVID-19) (NCT04318444), a randomized phase 2, phase 3 study. The primary outcome is Number of Participants With Symptomatic, Lab-confirmed COVID-19..

2. Clinical Question

This is a question about prevention after a household exposure, not treatment of established disease. The outcome definition requires both laboratory confirmation and self-reported symptoms. Confirmation could follow self-report of symptoms to the study or to a treating physician. The registry does not report whether the enrolled contacts had a confirmed exposure, how exposure timing was handled, or what procedures were used to establish infection and symptoms.

The stated purpose was to compare symptomatic secondary attack rates between randomized groups. However, the posted results are counts rather than rates, and no rate comparison is presented. The registry record also does not report a statistical analysis for the primary outcome. Therefore, the intended clinical question can be described, but the posted data do not quantify a comparative effect.

3. Trial Design

The registry identifies this as a randomized, parallel-group trial with quadruple masking and lists both Phase 2 and Phase 3. Columbia University is named as sponsor. The trial began on 2020-03-29, had a primary completion date of 2021-03-31, and is listed as terminated. The reason for termination is not reported in the registry record.

Participants were assigned to hydroxychloroquine or placebo. The registry describes the hydroxychloroquine schedule as two tablets (400mg) twice daily on day 1, then one tablet (200mg) twice daily on days 2-5. The placebo arm description repeats that schedule. The record does not explain the placebo's appearance, masking procedures, adherence monitoring, or how randomization was implemented. The assignment ratio is not reported in the registry record.

The registered enrollment is 10. The posted primary outcome data list 4 participants in the hydroxychloroquine group and 4 in the placebo group. These are separate registry fields; the record does not explain the relationship between the enrollment figure and the outcome-group counts. No reconciliation should be inferred from the posted information.

The primary endpoint was assessed from enrollment to 14 days post-enrollment date. It was defined as symptomatic COVID-19 confirmed by laboratory testing, with symptoms self-reported either to the study or to a treating physician. The record lists one outcome measure and no posted secondary outcomes. It does not report detailed eligibility criteria, recruitment procedures, follow-up completion, protocol deviations, or the reason enrollment ended at its recorded level.

The design features identify the intended comparison, but they do not establish the quality or completeness of the trial conduct. Details of allocation concealment, analysis populations, and handling of missing observations are not reported in the registry record. Those omissions matter when interpreting a small set of posted outcome counts.

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: Only 8 completers out of 10 enrolled/consented. 2 participants withdrew consent prior to initiating study intervention.

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.

Population: 2 out of 10 consented participants withdrew consent prior to initiating study intervention. Due to the low number of participants per group, baseline characteristics are presented as one arm to protect the confidentiality of the participants.

6. Primary Endpoint

The registry lists the primary outcome as Number of Participants With Symptomatic, Lab-confirmed COVID-19., measured over Date of enrollment to 14 days post-enrollment date.

The primary endpoint is the number of participants with symptomatic, laboratory-confirmed COVID-19 during the period from enrollment to 14 days post-enrollment date. A participant met the definition if COVID-19 infection was confirmed within that period after the participant reported symptoms to the study or to a treating physician.

This endpoint combines a disease event with symptoms and laboratory confirmation. Requiring confirmation distinguishes the endpoint from symptoms alone, while the symptom requirement makes it narrower than any laboratory-confirmed infection. It is suited to the registered question because the stated purpose concerned symptomatic secondary infection after household exposure.

As a count of participants with an event, the outcome is binary at the participant level: each participant either meets the endpoint definition or does not. A comparison would ordinarily describe the event occurrence in each randomized group and estimate the difference or another contrast, with uncertainty. The registry record does not provide a comparative estimate or its uncertainty. The posted event count is zero in each group, so the record supplies no observed event-positive participants for this endpoint.

The endpoint's interpretation depends on how symptoms, testing, and follow-up were ascertained. The registry definition names two routes for symptom reporting, but further details about testing access, timing, or outcome adjudication are not reported in the registry record. Whether participants with incomplete follow-up were counted as having no event or handled another way is also not reported.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry describes a randomized, parallel-group comparison with a binary primary endpoint. A binary endpoint records whether each participant meets a defined event criterion. Here the event is symptomatic, laboratory-confirmed COVID-19 in the specified follow-up period. The posted outcome is expressed as a count of participants with an event, not as a continuous measurement.

What does a two-arm comparison estimate?

In general, a two-arm analysis can estimate how event occurrence differs between assigned groups. Common summaries include an absolute difference in event proportions or a relative comparison. The choice depends on the design, the endpoint, and the analysis plan. The registry record does not name the planned effect measure or statistical test. It reports no analysis for the primary endpoint and marks a two-arm binary comparison as not estimable.

The event counts are zero in both groups. With no observed events, some relative measures have no defined value because their calculation relies on an event count in at least one group. An absolute comparison may be expressible under some approaches, but an estimate based on sparse observations has substantial uncertainty. That general explanation does not create an estimate for this trial. No comparative effect estimate is reported in the registry record.

Why does uncertainty matter?

A point estimate alone would not describe how uncertain a result is. A confidence interval is a range produced by a statistical procedure that reflects sampling uncertainty under its assumptions. With very sparse binary data, the interval can be broad, and different methods can give different results. The registry does not report a confidence interval for the primary endpoint. It also does not report a p-value or the method that would have been used to calculate one.

Random assignment is intended, in general, to make groups comparable on average, but it does not ensure balance in every small trial. Masking can reduce some forms of bias, yet the label “quadruple” does not specify who was masked or how masking was maintained. Those operational details are not reported in the registry record. Neither randomization nor masking makes an uninformative event pattern precise.

What can be inferred from the posted counts?

The counts document what the registry reports: zero primary events in each listed group. They do not establish that the true risks are equal, nor do they show that a difference is absent. There is no formal analysis to distinguish a real difference from chance variation. The record's indication that the comparison is not estimable should be read as a limit on the available analysis, not as proof of equivalence.

For a binary outcome, the denominator and definition of the analysis population are important. The group fields list 4 participants per arm, while the enrollment field lists 10. The registry does not explain this difference, and it does not say whether all randomized participants were included in the outcome count. Details of an intention-to-treat analysis, a per-protocol analysis, or another population are not reported. No population should be assumed.

What do planning illustrations mean?

In general, a sample-size calculation uses assumptions about the outcome rate, effect to detect, allocation, significance threshold, and desired power to plan a future study. A planning illustration is not evidence about the achieved power of a completed trial. The registry record does not report the trial's own sample-size justification or design assumptions. Any illustration based on observed data must be distinguished from the trial's original design, and it cannot repair the lack of an estimable comparative outcome in the posted record.

The record does not report interim analyses, a multiplicity adjustment, missing-data methods, or a detailed statistical analysis plan. Such methods should not be inferred from phase, masking, or randomization labels. In general, a statistical plan is needed to explain how outcomes, populations, and uncertainty measures were chosen. Without those details and without a reported analysis, interpretation should remain descriptive.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("Hydroxychloroquine Group", "Placebo Group")
n      <- c(4, 4)
events <- c(0, 0)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  Hydroxychloroquine Group  4  0  0.0  0.0  60.2  0.0  49.0  0.5  52.2
ARM  Placebo Group  4  0  0.0  0.0  60.2  0.0  49.0  0.5  52.2

11. Multiplicity, Missing Data and Analysis Populations

The registry lists one outcome measure and no posted secondary outcomes. It lists no analyses. The record does not report whether additional outcomes were collected, whether there was a prespecified hierarchy of hypotheses, or whether any adjustment for multiple comparisons was planned. With no reported comparative analysis, multiplicity procedures cannot be assessed from the registry record.

Missing-data handling is also not reported. The primary outcome groups each list 4 participants, while the enrollment field lists 10. The record does not explain the relationship between those figures, identify any participants with unknown outcome status, or say how missing follow-up was treated. It would be inappropriate to assume that every randomized participant had complete outcome data or that missing participants were counted as event-free.

The analysis population is not identified. The record does not say whether the analysis followed randomized assignment, included only participants who completed follow-up, or used another definition. No intention-to-treat or per-protocol result is posted. Such choices can affect interpretation, particularly when the number of observed events is small.

These gaps limit assessment of the reliability of the group counts and any comparison derived from them. They do not establish that data are missing or that an analysis was not performed outside the registry; they mean those details are not reported in the registry record. No conclusions about sensitivity analyses, missingness assumptions, or adjustment procedures can be drawn from the posted information.

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 5% reporting threshold.

The posted safety figures list zero serious adverse events and zero other adverse events in the hydroxychloroquine group, and zero serious and other adverse events in the placebo group. The registry lists 4 participants at risk for serious events and 4 at risk for other events in each group. No deaths are reported in either group, and no serious-event terms are listed.

These are descriptive registry counts. They do not show that no adverse events could occur, and they do not establish that the groups have equivalent safety. With few participants listed at risk, the absence of recorded events provides limited information about uncommon harms. The registry record does not report the safety collection procedures, event definitions beyond the serious/other categories, observation completeness, or a comparative safety analysis.

The “at risk” figures should be interpreted as the denominators recorded for these safety summaries, not as evidence of complete follow-up unless the record says so. The registry does not provide an explanation of how these safety denominators relate to the enrollment figure. No incidence rates, uncertainty intervals, or formal tests are reported. The safety counts therefore describe what was posted, while leaving substantial uncertainty about event rates and comparative safety.

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

The central finding in the registry record is that the listed primary outcome count is zero in each group. The outcome was symptomatic, laboratory-confirmed COVID-19 during the period from enrollment to 14 days post-enrollment date. The registry also lists zero serious and other adverse events in each group. These are limited descriptive observations, not a completed comparative assessment.

The posted results do not demonstrate that hydroxychloroquine reduces the primary outcome, and they do not demonstrate that it has no effect. The absence of a comparative estimate and uncertainty interval prevents a quantitative assessment of the difference between groups. Nor do zero events in the listed groups establish that the underlying event risk is zero. Sparse event data can be compatible with a range of possible underlying risks.

The record identifies a randomized, quadruple-masked design, but the posted results do not provide the detail needed to evaluate implementation, outcome completeness, or analysis choices. The enrollment field and outcome-group counts differ, and the record does not explain why. Missing-data handling and analysis populations are not reported. These limitations make it difficult to determine how fully the reported counts represent the randomized trial population.

The results should not be taken as proof of equivalence, proof of benefit, or proof of harm. Equivalence requires a suitable design and analysis with a prespecified margin; the registry record does not report such a framework. A failure to estimate a contrast is not the same as evidence that the contrast is zero.

The registry's terminated status is a trial-record fact, but the reason for termination is not reported. It should not be used to infer why the trial stopped or what its investigators concluded. The most defensible reading is narrow: the record documents the planned comparison and reports no primary events among the participants listed in either outcome group, while providing no estimable comparative result.

14. Why This Trial Matters Statistically

This trial is statistically instructive because randomization and masking do not guarantee that a study will produce an informative comparison. The registry identifies a randomized, quadruple-masked, parallel-group design, yet the posted primary endpoint has zero events in both listed groups and no comparative analysis is available. The example illustrates the distinction between a sound design label and the amount of information contained in observed outcomes.

Binary outcomes are especially challenging when events are sparse. General statistical methods rely on the pattern of events and the number of observations to estimate group differences and uncertainty. When no events are recorded in either group, certain comparisons cannot be estimated in the usual way, while other summaries may be too imprecise to support a useful conclusion. The registry's notation that the two-arm comparison is not estimable makes this limitation explicit.

The record also illustrates why denominators and population definitions matter. The enrollment field lists 10, but each outcome group lists 4 participants. Without an explanation, readers cannot determine how those figures relate, whether outcome data were incomplete, or which participants were included in the posted analysis. Statistical interpretation depends on knowing who contributed data and how missing outcomes were handled.

Finally, the trial shows why a planning calculation should not be confused with an observed result. In general, sample-size planning relies on assumptions and does not establish what a completed study has learned. When the observed endpoint is too sparse for a comparison, a planning illustration cannot substitute for an effect estimate or resolve uncertainty. The registry's limited outcome reporting makes transparency about these distinctions particularly important.

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 limitation is that the registry posts zero events in both outcome groups and no comparative analysis. No effect estimate, confidence interval, or p-value is reported. The binary comparison is marked as not estimable, so the available endpoint information cannot quantify a difference between assigned groups.

The outcome-group counts are 4 in each arm, whereas the registered enrollment is 10. The registry does not explain the difference. It is not possible to determine whether the figures reflect different analysis populations, missing outcomes, exclusions, or another registry-record issue. Follow-up completeness, missing-data handling, and the analysis population are not reported.

The record lists the primary endpoint and its 14-day assessment period, but details of symptom ascertainment, laboratory testing, outcome adjudication, and follow-up procedures are not reported. Such details can affect whether events are detected and classified consistently. The registry also does not provide the rationale for termination, the sample-size justification, interim-analysis procedures, or a statistical analysis plan.

The design is labeled randomized and quadruple-masked, but operational details of randomization and masking are absent. The registry lists one outcome measure and no secondary outcomes or analyses; it does not explain whether other outcomes were collected or how multiplicity was addressed. Safety data are also sparse and lack details about ascertainment and completeness.

These limitations concern what can be assessed from the registry record. They do not prove that study procedures or analyses were inadequate; they mean that the posted information does not permit a fuller evaluation. Conclusions should remain restricted to the recorded counts and the absence of a reported comparative result.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Hydroxychloroquine Post Exposure Prophylaxis for Coronavirus Disease (COVID-19) — NCT04318444.
• Registry record last updated 2024-09-25; status last verified 2024-08.
• Results first posted 2024-09-25.
• 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 describes a randomized, parallel-group, quadruple-masked Phase 2 and Phase 3 trial sponsored by Columbia University. It compared hydroxychloroquine with placebo as post-exposure prophylaxis for household contacts of known or suspected COVID-19 patients. The recorded start date is 2020-03-29, the primary completion date is 2021-03-31, and the status is terminated.

The primary outcome was symptomatic, laboratory-confirmed COVID-19 from enrollment to 14 days post-enrollment date. The posted count is zero in the hydroxychloroquine group and zero in the placebo group; each outcome group lists 4 participants. The registry lists enrollment of 10 and does not explain how it relates to the outcome-group counts.

No analyses or secondary outcomes are posted. The registry indicates that a two-arm comparison for the binary endpoint is not estimable. Safety summaries list zero serious events, zero other events, and zero deaths in each group. The reason for termination, analysis population, missing-data procedures, sample-size justification, and detailed statistical plan are not reported in the registry record.