COVID-19 PrEP HCW HCQ Study: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for COVID-19 PrEP HCW HCQ Study (NCT04354870), a non-randomized phase 2 study. The primary outcome is Change From Baseline in the Number of Participants With Seroconversion to SARS-CoV-2 at 1 Month.

2. Clinical Question

The clinical question concerns whether seroconversion, measured through antibody titers, differed between workers who chose HCQ and workers who chose not to take it. The primary endpoint is described as a change from baseline in the number of participants with seroconversion at 1 month. The record does not report the primary outcome counts or a between-group estimate, so the direction and precision of any observed difference cannot be assessed from the registry information summarized in the record.

Because participants chose whether to take HCQ, the comparison is not randomized. The groups may therefore differ in ways that affect exposure, testing, or the likelihood of seroconversion, apart from any association with the medication. The registry record does not report whether or how such differences were measured or addressed. The question is consequently a comparison of participant groups formed by choice, not a randomized estimate of an intervention effect.

3. Trial Design

The study is listed as a Phase 2, parallel-group, non-randomized trial with no masking. NYU Langone Health is the sponsor, and the condition is COVID-19. The record lists a start date of 2020-04-03, a primary completion date of 2020-09-15, and a completed status. The recorded enrollment is 130.

The study description says that health care workers at high occupational risk could choose whether to take HCQ. The HCQ group is described as workers choosing to receive the medication, while the control group is described as workers choosing not to receive it. The brief summary refers to 350 participants, with approximately 300 in the HCQ group and approximately 50 in the control group. This planned description does not match the recorded enrollment of 130; the reason for that difference is not reported in the registry record. The outcome risk sets list 83 participants in the HCQ group and 47 in the control group, but the record does not explain how those sets relate to enrollment or to missing outcome data.

Study visits were planned at screening or enrollment and at 30 day, 60 day, and 90 day timepoints, with questionnaires and dried blood spot collection described for all timepoints. The primary endpoint is assessed at 1 month, which corresponds to the stated first follow-up timepoint. How the non-randomized comparison was analyzed, whether adjustments were planned, and whether any interim analyses were conducted are not reported in the registry record. The record lists no analyses.

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:

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 Change From Baseline in the Number of Participants With Seroconversion to SARS-CoV-2 at 1 Month, measured over Baseline to 1 month post-baseline.

The primary endpoint is the number of participants with seroconversion to SARS-CoV-2 at 1 month, described as a change from baseline. Seroconversion is defined in the registry record as a binomial endpoint: antibody titers must show a four-fold increase compared with baseline. The endpoint therefore classifies participants according to whether they meet that threshold, rather than measuring a continuous antibody change as the primary outcome.

General statistical context: a binary endpoint can be summarized as the number of participants meeting the definition and the number assessed. Comparing such counts between groups can describe how frequently the event occurred, while an estimate of the between-group contrast and its confidence interval would convey the size and precision of that difference. Those estimates are not reported in the registry record.

Antibody-based seroconversion is a measurable response to infection, but it is not itself a direct measure of every clinically important aspect of COVID-19. The registry record does not report additional details about the assay, threshold validation, testing schedule beyond the stated timepoints, or handling of participants without an available baseline or follow-up measurement. Those details would help readers judge how consistently the endpoint could be classified.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The endpoint is binary: each participant either meets the seroconversion definition or does not. The definition uses a four-fold increase in antibody titers compared with baseline. The primary outcome is framed as the change from baseline in the number of participants meeting that definition at 1 month. The registry record does not report the primary outcome counts or a statistical comparison.

What does a binary outcome count represent?

General statistical context: a count of participants with an event is interpretable only alongside the number assessed. The event count describes how many participants met the endpoint, while the denominator gives the group size used to contextualize that count. If participants lack measurements, the number assessed may differ from the number enrolled. The registry record lists outcome risk sets of 83 for the HCQ group and 47 for the control group, but does not report how missing primary endpoint measurements were handled or whether these risk sets are the primary endpoint denominators.

A difference in counts alone is not generally enough to compare groups of different sizes. Statistical comparisons may use event proportions or other measures, with an uncertainty interval. No primary event counts, comparison estimate, uncertainty interval, or p-value are reported in the registry record, so none can be interpreted for this outcome.

Why does allocation matter?

General statistical context: random allocation is intended to make treatment groups comparable on average with respect to both measured and unmeasured characteristics. In a non-randomized study, participants may differ because of their choices or circumstances. Here, workers chose whether to take HCQ. Differences between those who chose HCQ and those who did not could influence infection risk or the likelihood of testing, independently of the intervention. The registry record does not report a statistical adjustment strategy or enough participant characteristics to assess the extent of this concern.

The study is also listed as unmasked. General statistical context: when participants and study personnel know group assignment, behavior, follow-up, reporting, or outcome assessment can potentially differ by group. The record does not describe safeguards for outcome assessment or whether seroconversion testing was interpreted without knowledge of group. This does not demonstrate that ascertainment differed; it identifies information that would be useful for judging the comparison.

How should uncertainty be understood?

General statistical context: a confidence interval gives a range of values compatible with the observed data under the assumptions of the analysis. Its width reflects statistical precision, but it does not account automatically for systematic differences between non-randomized groups, measurement error, or missing data. Because the primary estimate and interval are not reported in the registry record, the precision of the seroconversion comparison cannot be assessed.

The registry describes sample-size illustrations as planning exercises based on an observed effect, with a two-sided significance level of 0.05, equal allocation, and a normal approximation. These are planning assumptions, not evidence about the actual trial analysis. Sample-size illustrations are not observed power, and they do not replace a reported effect estimate and confidence interval. The registry record does not report the trial's own sample-size justification or a detailed statistical analysis plan.

What can the safety counts tell readers?

The listed safety fields include participants at risk and participants with serious adverse events, other adverse events, and deaths. They describe counts, not adjusted comparisons. General statistical context: with uncommon outcomes, small counts can leave substantial uncertainty about the underlying event frequency. A zero count means no event was recorded in the listed group and risk set; it does not establish that the event is impossible. The record does not report uncertainty intervals for these counts, detailed event definitions, or the full set of serious adverse event terms.

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("HCQ Group", "Control Group")
n      <- c(56, 24)
events <- c(4, 2)
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  HCQ Group  56  4  7.1  2.0  17.3  2.8  17.0  2.9  17.0
ARM  Control Group  24  2  8.3  1.0  27.0  2.3  25.8  2.5  26.0

11. Multiplicity, Missing Data and Analysis Populations

The registry record lists three outcome measures and 0 analyses. It does not report a multiplicity strategy, such as whether one outcome was designated as the sole confirmatory test or whether any adjustment was planned for examining multiple outcomes. Without that information, the implications of testing or interpreting several outcomes cannot be evaluated. The primary endpoint is identified, but the record does not report its numerical result or an analysis result.

Missing data are also not described. The record does not report how many participants completed each visit, how many had baseline and follow-up antibody measurements, why any measurements were missing, or how missing observations were handled in the primary analysis. The listed outcome risk sets are 83 in the HCQ group and 47 in the control group, but the record does not explain their relationship to the recorded enrollment or to endpoint availability. It would be inappropriate to assume that all enrolled participants contributed to the primary endpoint.

The registry record does not report analysis populations, such as whether the primary analysis included all enrolled participants, only those with available measurements, or another defined group. It also does not report whether participants were analyzed according to their choice at enrollment if their medication use later changed. These details affect how the analysis population corresponds to the original study question.

General statistical context: multiple outcome testing can increase the chance of finding an apparently notable result by chance, while missing-data methods can affect both the estimate and its uncertainty. These concerns cannot be resolved by the outcome titles or counts alone. The analysis plan, population definitions, multiplicity approach, and missing-data procedures are not reported in the registry record.

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 0% reporting threshold.

Most frequent non-serious adverse event terms

The safety listing reports 0 serious adverse events among 83 participants at risk in the HCQ group and 0 among 47 participants at risk in the control group. It also lists 0 deaths in each group. These are observed counts for the listed groups and risk sets. A zero count should not be read as proof that an event cannot occur, nor does it establish equivalent safety between the groups.

For other adverse events, the listing reports 8 among 83 participants at risk in the HCQ group and 0 among 47 in the control group. The record does not give event terms for these other adverse events in the registry record for this analysis. The secondary outcome describes adverse events or serious adverse events possibly, probably, or definitely related to HCQ upon study termination, but the listed other-event count alone does not establish that those events were judged related to HCQ. The record does not report the attribution details for the 8 events.

The serious-event term listing is empty, so serious adverse event terms are not reported in the registry record. No uncertainty intervals or comparative safety analyses are reported. Because allocation was non-randomized and participants chose their group, the counts are descriptive and do not isolate an effect of HCQ from differences between participants.

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

The registry record describes a study comparing health care workers who chose HCQ with workers who chose not to take the study medication. It defines a primary seroconversion endpoint and reports some safety counts. It does not report the primary outcome result, a between-group estimate, or its precision. The central efficacy question therefore remains unanswered by the results available in the registry record.

The design also limits what a group comparison could establish. Participants selected their treatment group, and the study was unmasked. Any observed difference could reflect the intervention, differences between participants, or other factors; the registry record does not report an analysis that addresses those possibilities. The descriptive safety counts likewise do not establish that one group had a higher or lower underlying risk.

General statistical context: an observed association and a causal treatment effect are not interchangeable. A causal interpretation requires assumptions and design or analysis features that address alternative explanations. Randomization can support that goal, but this study did not use random allocation. No adjustment methods are reported in the registry record.

The outcome itself is a threshold-based antibody measure at 1 month, not a complete account of infection, illness severity, or other clinical outcomes. The registry record does not report enough detail to assess assay performance or missing measurements. The study's planned later visits do not supply later outcome results in the information.

Accordingly, the record supports a description of the study question, design, endpoint definition, and listed adverse event counts. It does not support a conclusion about whether HCQ prevented infection or changed the probability of seroconversion, nor about the size or precision of any such effect. Those results and the analysis details are not reported in the registry record.

14. Why This Trial Matters Statistically

This study illustrates a recurring statistical challenge in clinical research: a biologically measurable endpoint does not by itself ensure that a comparison can be interpreted causally. Seroconversion is defined using antibody titers and a threshold relative to baseline, making the primary endpoint a participant count. But the comparison is non-randomized, with group membership determined by participant choice. That distinction matters when interpreting any difference between groups.

The record also makes clear why enrollment, outcome denominators, and analysis populations need separate descriptions. It records enrollment of 130, while the study summary describes 350 planned participants and approximate group sizes. The outcome risk sets list 83 and 47. The registry record does not explain the relationship among these figures. A reader needs that information to understand who was enrolled, followed, and included in each outcome analysis.

Another statistical lesson is the difference between a defined endpoint and a reported result. The endpoint is described in detail, but its numerical outcome and comparative estimate are not reported in the registry record. Without those results, there is no way to assess the observed effect or its precision. A complete interpretation also depends on a prespecified analysis, missing-data handling, and clarity about multiple outcomes, none of which is reported in detail.

The safety counts offer descriptive information but require similar care. Zero serious adverse events in the listed groups does not establish equal risk, while other adverse event counts do not establish treatment attribution. Taken together, the record is useful for understanding study structure and the limits of the available evidence, while demonstrating why complete outcome reporting and transparent methods are essential for statistical interpretation.

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 most immediate limitation is that the primary seroconversion result is not reported in the registry record. There are no primary event counts, comparative estimates, confidence intervals, or p-values in the registry record for this analysis. The record therefore does not permit assessment of effect direction, magnitude, or precision.

Allocation was non-randomized, and participants chose whether to take HCQ. The groups may have differed in relevant ways, but baseline characteristics and adjustment procedures are not reported in the registry record. The study was also unmasked; whether outcome assessment was protected from knowledge of group assignment is not reported.

The enrollment information is difficult to reconcile: the record lists enrollment of 130, while the brief summary describes 350 participants and approximate group sizes. The outcome risk sets are 83 and 47, and the record does not explain the difference between those figures and enrollment, or how many participants had primary endpoint measurements. Reasons for missing data, visit completion, and the handling of incomplete measurements are not reported.

Other details important to interpretation are also absent. The registry record does not report the sample-size justification, detailed statistical analysis plan, analysis populations, multiplicity procedures, interim analyses, or methods for addressing possible differences between groups. It does not provide assay details beyond the seroconversion definition. The serious adverse event term listing is empty, and the attribution of other adverse events is not reported. These limitations restrict both efficacy interpretation and the conclusions that can be drawn from the safety counts.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: COVID-19 PrEP HCW HCQ Study — NCT04354870.
• Registry record last updated 2022-12-14; status last verified 2021-10.
• Results first posted 2021-10-06.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

19. Record Summary

The registry record describes a completed Phase 2, parallel-group, non-randomized, unmasked study sponsored by NYU Langone Health. It evaluated HCQ pre-exposure prophylaxis among health care workers at high occupational risk of SARS-CoV-2 exposure. Participants chose whether to take the study medication. Recorded enrollment is 130, while the study summary describes 350 participants; the reason for this difference is not reported in the registry record.

The primary endpoint was the number of participants with seroconversion at 1 month, defined as a four-fold increase in antibody titers compared with baseline. The primary outcome result and statistical comparison are not reported in the registry record. Listed safety counts include 0 serious adverse events in each group, 8 other adverse events in the HCQ group and 0 in the control group, and 0 deaths in each group. The record lists 0 analyses and does not report a detailed analysis plan.