Treating COVID-19 With Hydroxychloroquine (TEACH): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Treating COVID-19 With Hydroxychloroquine (TEACH) (NCT04369742), a randomized phase 2 study. The primary outcome is Percent of Participants With SAE Through Day 30.

2. Clinical Question

The clinical question was whether hydroxychloroquine, compared with placebo, differed in outcomes among hospitalized adults with COVID-19. The registry record describes the trial as randomized, double-blind and placebo-controlled. Its study summary says the trial was intended to address uncertainty about HCQ in human infection; the summary also describes planned clinical and exploratory outcomes.

The distinction between the study summary and the posted results matters. The summary names a severe-disease-progression composite at the 14-day post-treatment evaluation as the planned primary endpoint. The primary result posted in the registry record is instead the percentage with a serious adverse event through day 30. The record does not explain this difference, so these endpoints should not be treated as interchangeable. The result should be read as a comparison of the posted day-30 safety measure, alongside the specific secondary outcomes with results, rather than as a complete answer to every question in the study summary.

3. Trial Design

The registry record describes a phase 2, randomized, double-blind, parallel-group trial sponsored by NYU Langone Health. The listed arms were hydroxychloroquine and placebo. The study summary describes a five-day course of HCQ or placebo and a planned 1:1 allocation. The record lists enrollment of 128 participants. The two posted primary-result groups include 67 participants in the hydroxychloroquine arm and 61 in the placebo arm.

The study summary describes the intended population as adults hospitalized with laboratory-confirmed COVID-19. It identifies ICU admission or ventilation at enrollment, prior HCQ therapy, and baseline prolonged qTC as notable exclusions. These details describe the intended trial population and eligibility; the registry record does not provide participant-level information showing how many people met each criterion.

The record lists a start date of 2020-04-15, a primary completion date of 2021-04-01, and a terminated status. The summary describes a planned enrollment of 626, whereas the registry enrollment is 128. The reason for termination, the timing of the decision, and the circumstances behind the difference between planned and listed enrollment are not reported in the registry record.

Several outcome descriptions require care in interpretation. The study summary names a day-14 severe-disease-progression composite as its primary endpoint, while the posted primary result is serious adverse events through day 30. The record does not explain the change or discrepancy. The reported result therefore speaks directly to the posted endpoint; it does not establish what analysis was planned or performed for the summary's composite outcome. The record also lists 27 outcome measures, with 23 secondary outcomes posted, but does not report analyses for those outcomes as sponsor-reported 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 Percent of Participants With SAE Through Day 30, measured over 30 days.

The posted primary endpoint is the percentage of participants with a serious adverse event through day 30. It is a binary outcome: each participant either had at least one qualifying event during the stated period or did not. The record defines the reported value as the number of patients with the outcome divided by the number of patients, multiplied by 100.

Binary endpoints are useful for describing how often a specified event occurred in each group. Comparing the group percentages gives a direct summary of event frequency, while a confidence interval around a between-group estimate indicates how much uncertainty remains about the difference. A safety endpoint can provide information about serious events recorded during follow-up, but it is not the same as an endpoint measuring disease progression or treatment benefit.

The hydroxychloroquine group had a posted value of 14.9%, and the placebo group had 16.4%. The registry record's study summary describes a different planned primary endpoint: a composite of severe disease progression assessed at the 14-day post-treatment evaluation. That composite was described as including death, ICU admission, mechanical ventilation, ECMO, and/or vasopressor requirement. The record does not explain why the primary result posted is a day-30 serious adverse event measure rather than that day-14 composite. Consequently, the interpretation of the posted primary result should remain specific to serious adverse events through day 30.

7. Results Posted in the Registry

• Hydroxychloroquine: 14.9% (10/67) versus 16.4% (10/61) on control; risk difference -1.5 percentage points (95% CI -14.1 to 11.1), risk ratio 0.91 (0.41–2.04); unadjusted two-proportion z-test p = 0.819.

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.

Hospital Length of Stay

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • Hydroxychloroquine: mean 9.75 versus 6.80 on control; difference 2.95 (95% CI 0.07 to 5.83), standardized effect d = 0.35; z-test p = 0.045.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically higher for Hydroxychloroquine than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Percent of Subjects With Q-T Interval, Corrected (qTC) Prolongation at End of Treatment (EOT)

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Hydroxychloroquine: 24.4% (10/41) versus 12.8% (5/39) on control; risk difference 11.6 percentage points (95% CI -5.2 to 28.4), risk ratio 1.90 (0.71–5.07); unadjusted two-proportion z-test p = 0.185.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Hydroxychloroquine and the control. The interval gives the range of effects that remain compatible with the data.

Percent of Patients Who Resulted in Mortality

• Endpoint type: Binary: the percentage of participants with the event (counts rebuilt from the posted percentages and group sizes).
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Hydroxychloroquine: 13.2% (7/53) versus 12.0% (6/50) on control; risk difference 1.2 percentage points (95% CI -11.6 to 14.0), risk ratio 1.10 (0.40–3.05); unadjusted two-proportion z-test p = 0.854.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Hydroxychloroquine and the control. The interval gives the range of effects that remain compatible with the data.

The posted secondary outcomes address several distinct aspects of the trial: duration of hospitalization, QTc prolongation at end of treatment, and mortality through 30 days. Their observed comparisons do not form a uniformly precise pattern. One outcome has an unadjusted p-value below 0.05, while the other two have confidence intervals that cross zero and p-values that do not clearly distinguish the groups.

These measures are related to the primary result only in the broad sense that they describe outcomes observed in participants assigned to the two trial arms. They are not interchangeable measures. Serious adverse events, hospital length of stay, QTc prolongation, and mortality capture different events, use different follow-up periods or denominators, and may have different sources of variation. A result for one cannot be used as a substitute for the others or for the severe-disease-progression composite described in the study summary.

The secondary findings do not resolve the uncertainty in the primary comparison. Nor do they establish that the treatment groups are equivalent on outcomes whose intervals include zero. The hospital-stay comparison is noteworthy as an unadjusted result, but it was one of multiple outcomes listed in the registry record. When many outcomes are examined, an isolated low p-value can occur by chance more often than in a single prespecified test. The record does not report a multiplicity adjustment or a hierarchy that would establish which secondary outcome should carry confirmatory weight.

The secondary outcomes are therefore best treated as descriptive and exploratory evidence. Interpretation should account for the endpoint-specific denominators, the lack of reported sponsor analyses, and uncertainty about missing observations and analysis populations. They can identify patterns for further evaluation, but they do not, on their own, settle the interpretation of the posted primary result or answer the distinct primary question described in the study summary.

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

• Hospital Length of Stay
• Days of Fever
• Days of Non-invasive Ventilator Use
• Days of Non-rebreather Mask Oxygen Supplementation
• Number of Participants With Mild, Moderate, and Severe Scores on Cytokine Release Syndrome (CRS) Grading Scale
• Percent of Subjects With Q-T Interval, Corrected (qTC) Prolongation at End of Treatment (EOT)
• Percent of Patients Who Resulted in Mortality
• Percent of Participants Who Required ICU Admission

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is a binary percentage: the proportion of participants with a serious adverse event through day 30. For a binary outcome, each participant contributes to an event count or a non-event count. The percentage describes the event frequency in a group. A risk difference compares those frequencies by subtracting one group’s risk from the other; its sign indicates which group had the higher observed risk, given the comparison’s stated direction. The reported group percentages are 14.9% for hydroxychloroquine and 16.4% for placebo.

The registry comparison is an unadjusted recalculation from the posted summary statistics. It reports a risk difference, a confidence interval, and an unadjusted p-value of 0.819. “Unadjusted” means the comparison is not described as controlling for other variables or for the number of outcomes examined. The registry record reports no discrepancy between the recalculation and the posted values. No sponsor-reported analysis is listed for the secondary outcomes. The registry record does not report a detailed analysis plan, including model specifications, covariate adjustment, or a prespecified multiplicity procedure.

A confidence interval gives a range of values compatible with the data under the method’s assumptions. For a risk difference, a confidence interval that includes zero does not clearly distinguish an increase from a decrease in risk. It does not prove that there is no difference. The primary interval extends across zero, and it is broad enough to leave substantial uncertainty about the direction and size of the difference. The same general interpretation applies to the intervals for QTc prolongation and mortality, which also cross zero.

The hospital-length-of-stay outcome is continuous and is reported as a mean with a standard deviation for each group. A mean difference compares the group averages. The standard deviation describes variation among participant values within a group; it is not the uncertainty in the estimated mean difference. For this outcome, the reported means are 9.75 days and 6.8 days, and the mean difference is 2.95 days. The unadjusted p-value is 0.045. Because the outcome is a length of stay, its distribution could be asymmetric or influenced by unusually long stays; the registry record does not report whether an alternative analysis was considered. This is a general concern for continuous outcomes, not a claim about the observed data.

For binary endpoints, relative measures such as a risk ratio or odds ratio compare groups on a multiplicative scale, whereas the risk difference is an absolute-scale comparison. These measures can describe the same event data from different perspectives. The posted comparisons include relative measures for the primary endpoint and selected secondary outcomes, but their intervals are wide. The clearest interpretation should remain anchored to the observed percentages, the absolute comparison, and the uncertainty rather than any one scale alone.

Some secondary outcomes have fewer participants contributing than the primary endpoint. The QTc result includes 41 participants in the hydroxychloroquine group and 39 in the placebo group; the mortality result includes 53 and 50, respectively. Different available denominators can affect precision and make comparisons across outcomes less direct. The reasons for these endpoint-specific denominators, the handling of missing observations, and whether an intention-to-treat or other analysis population was used are not reported in the registry record.

Multiplicity is the statistical issue that arises when several outcomes or comparisons are tested. If many tests are considered, the chance of at least one apparently unusual result can increase. The registry record lists 27 outcome measures and 23 posted secondary outcomes, but it does not report whether a hierarchy, adjustment, or other multiplicity strategy was used. An isolated unadjusted p-value should therefore be interpreted cautiously, especially when several outcomes are being considered. The p-value for hospital length of stay does not by itself establish a confirmatory finding.

The study summary describes a planned day-14 composite of severe disease progression, but that is not the endpoint of the posted primary result. The posted primary endpoint is serious adverse events through day 30. The record does not report the reason for this difference, the analysis population for the composite, or a sponsor-reported analysis of it. Statistical conclusions should be tied to the endpoint actually represented by the posted numbers, without assuming that the summary and results refer to the same analysis.

11. Confidence Intervals and Precision

The primary estimate is not precise enough to identify a clear direction of difference. The risk-difference confidence interval extends from a negative value to a positive value, and the unadjusted p-value is 0.819. In general, a confidence interval expresses the uncertainty around an estimate under the assumptions of the method. When an interval includes zero for a difference, the data do not clearly distinguish which group has the higher underlying event frequency. This does not establish equality.

The posted primary percentages, 14.9% and 16.4%, are close as observed values, but the interval is much more informative about precision than the closeness of the percentages alone. Its span allows meaningfully different underlying comparisons to remain compatible with the data. The registry record lists 67 participants in the hydroxychloroquine group and 61 in the placebo group for this endpoint. Those group sizes, together with the event counts and statistical method, influence the uncertainty; the record does not provide a formal precision target for the primary comparison.

Precision is also limited for the secondary outcomes. The interval for hospital length of stay is positive, but ranges from 0.071 to 5.829 days, indicating uncertainty about the size of the difference. The QTc and mortality risk-difference intervals include zero and extend in both directions. Their posted denominators are smaller than the primary endpoint denominators, and the record does not explain the missing outcome-specific observations.

Intervals should be interpreted as ranges of values compatible with the observed data and method, not as a guarantee that the true value lies within the displayed limits. They also do not account automatically for unreported design features, missing-data choices, or multiple outcome testing. Here, the interval and endpoint-specific sample sizes support restrained interpretation rather than a precise ranking of the groups across outcomes.

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 planning illustrations give participants per arm for the binary and continuous outcomes discussed in the registry record; they are not totals across both arms and are not total-event counts. The assumptions listed for these illustrations are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is set to the stated fraction of the observed effect. The original trial’s own sample-size assumptions and justification are not reported in the registry record.

The 80% power column describes the number of participants per arm that the illustration estimates would be needed to have an 80% chance of detecting the specified effect under its assumptions. The 90% power column gives the corresponding planning number for a 90% chance. Power is a property of a design and an assumed effect, not a measure of whether a completed trial’s result is true or important. These figures are not observed power.

Rows planning for 100%, 75%, and 50% of the observed effect show how the target effect changes the planning requirement. If the effect to detect is smaller, the groups generally need more participants to distinguish it from random variation at the stated significance level and power. This relationship is visible across the listed planning scenarios: the smaller-effect scenarios require larger per-arm numbers.

These numbers are planning illustrations based on observed effects, not a reconstruction of the trial’s intended design. The trial record describes planned enrollment of 626 in its study summary and lists enrollment of 128, but the reason for the difference and the original design rationale are not reported. The illustrations should not be read as evidence that the original target was chosen from these calculations or that any particular observed result was adequately powered.

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 = 0.164; treatment proportion p₁ = 0.149; average p̄ = 0.157.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.5140 = 1.0073.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.5138 = 0.4325.
4. n = (1.0073 + 0.4325)² / (-0.0147)² = 9,618.4, rounded up to 9,619 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 <- 0.163934426229508                          # control proportion
p1 <- 0.149253731343284                          # 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(4369742)
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 = 9,619 (80% power), 12,876 (90% power)
 75% of observed effect: n per arm = 17,264 (80% power), 23,110 (90% power)
 50% of observed effect: n per arm = 39,208 (80% power), 52,488 (90% power)
power.prop.test: 9618.4 
Simulated power at n = 9,619 per arm: 0.799 (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. Multiplicity, Missing Data and Analysis Populations

The registry record lists 27 outcome measures and reports 23 secondary outcomes as posted. Testing or interpreting multiple outcomes raises a multiplicity concern: even when there is no underlying difference for any outcome, considering many comparisons makes it more likely that at least one result will appear unusual by chance. This is a general statistical issue, not evidence that a particular reported finding is false.

The record does not report a prespecified hierarchy for testing the outcomes, a family of outcomes for formal inference, or a correction for multiple comparisons. It also does not report whether the hospital-length-of-stay result was intended as confirmatory. Its unadjusted p-value should consequently be read as a nominal result, not as a multiplicity-adjusted conclusion. The other secondary p-values also do not provide a complete account of the trial’s evidence across all outcomes.

Missing-data handling is not reported in the registry record. The posted denominators differ by endpoint: the primary serious-adverse-event result includes 67 and 61 participants, while the QTc and mortality outcomes include smaller groups. The record does not say why observations are absent, whether participants lacked an assessment, or how any missing outcomes were treated. It would be inappropriate to assume that the missingness was random or that a particular imputation method was used.

The analysis populations are also not reported in the registry record. The record does not state whether results used all randomized participants, participants who received treatment, participants with available endpoint data, or another defined population. Nor does it report a detailed statistical analysis plan, covariate adjustments, interim analyses, or a sample-size justification for the original trial. These omissions limit assessment of how closely the posted results correspond to planned analyses and whether the reported comparisons account for the trial’s design and outcome structure.

15. 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 serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: death (10.4%), acute renal failure (1.6%), diverticulitis (1.6%), hospice care (1.6%), hospitalization (1.6%).

For serious adverse events through day 30, the posted percentages were 14.9% among 67 participants in the hydroxychloroquine group and 16.4% among 61 participants in the placebo group. The adverse-event record lists 10 participants affected by a serious adverse event in each group. These values describe the recorded events in the posted groups; they do not by themselves establish a difference in underlying safety risk.

The adverse-event record also lists deaths of 7 in the hydroxychloroquine group and 6 in the placebo group. A separate 30-day mortality outcome reports 13.2% and 12.0%, respectively, using endpoint denominators of 53 and 50. These are distinct summaries with different denominators. The registry record does not explain the difference in denominators, so the counts and percentages should be read in their stated contexts.

Among the listed serious-event terms, Death is reported as 10.4%; Acute renal failure, Diverticulitis, Hospice care, Hospitalization, and Hyperkalemia are each reported as 1.6%. Event categories may overlap, and the record does not provide enough detail to treat these terms as mutually exclusive or to infer causal attribution. These safety figures describe what was recorded; they do not establish that an event was caused by either assigned treatment.

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

The posted results describe outcomes among participants in the two randomized groups, with the main reported comparison concerning serious adverse events through day 30. The observed percentages were 14.9% with hydroxychloroquine and 16.4% with placebo. The interval for the risk difference crosses zero, and the unadjusted p-value is 0.819. These findings leave substantial uncertainty about the direction and size of any underlying difference.

That uncertainty is not evidence that the groups are the same. A non-significant result can arise when there is no meaningful difference, but it can also arise when available data are too imprecise to distinguish a difference from chance variation. The record does not report an equivalence or non-inferiority design, a margin for such a conclusion, or an analysis that would support it. The posted primary result therefore does not establish equivalence.

The result is also not a direct test of the day-14 severe-disease-progression composite described in the study summary. The posted primary endpoint is a day-30 serious adverse event measure. The record does not explain why those endpoints differ. Conclusions about the posted safety endpoint should not be extended to the planned composite or to all clinical outcomes.

The secondary results add descriptive information but do not remove the limitations in the primary comparison. They involve distinct outcomes and differing denominators, and multiple testing is not addressed in the registry record. In particular, the hospital-length-of-stay result has an unadjusted p-value of 0.045, but it should not be treated as definitive evidence in isolation when the multiplicity strategy is unreported.

The trial is listed as terminated and enrolled 128 participants, while its study summary describes a planned enrollment of 626. The reason for termination and the reason for the difference are not reported in the registry record. These registry entries, together with the endpoint discrepancy and unreported analysis details, counsel against treating the posted results as a complete resolution of the intended research question. They support a narrow reading of the observed outcomes and their uncertainty, not a general claim about treatment efficacy or lack of efficacy.

17. Why This Trial Matters Statistically

This trial illustrates why endpoint identity matters when reading a clinical trial record. The study summary describes a day-14 severe-disease-progression composite, while the posted primary result is serious adverse events through day 30. Those outcomes differ in definition and follow-up. Statistical interpretation must follow the endpoint attached to the actual result, rather than assume that the posted estimate answers the summary’s question.

It also demonstrates the importance of precision alongside point estimates and p-values. The primary group percentages are close, but the confidence interval crosses zero and leaves meaningful uncertainty about the direction and size of a difference. A p-value alone cannot establish equivalence, and a result that does not meet a conventional significance threshold does not prove that no difference exists.

The registry record also makes visible the distinction between planned and realized enrollment. The summary describes planned enrollment of 626, whereas the listed enrollment is 128 and the trial status is terminated. Since the reason for termination and the original design rationale are not reported, those details cannot be used to infer why enrollment ended or how the trial’s operating characteristics were affected. Planning illustrations based on observed effects are not a substitute for the original design assumptions.

Finally, the secondary results show why multiplicity and endpoint-specific denominators matter. Different outcomes can have different observed patterns and different precision, and a nominally low p-value among multiple outcomes warrants caution when the analysis strategy is not reported. Taken together, these features make transparent reporting of endpoint definitions, analysis populations, missing data, and multiplicity especially important for interpreting clinical trial results.

18. 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.

Several limitations are specific to the information recorded for this trial. The most important is the difference between the primary endpoint described in the study summary and the posted primary result. The summary names a day-14 severe-disease-progression composite, while the posted result concerns serious adverse events through day 30. The record does not explain the difference, so it is unclear from the registry record how the posted result relates to the intended primary question.

The record lists enrollment of 128, compared with planned enrollment of 626 in the study summary, and gives the trial status as terminated. The reason for termination, timing of termination, and reasons for the enrollment difference are not reported. The original sample-size justification and operating assumptions are also not reported; the planning illustrations based on observed effects cannot fill that gap.

The registry record does not report a complete statistical analysis plan, analysis populations, handling of missing data, or multiplicity procedures. Participant counts vary across outcomes, but reasons for the varying denominators are not described. The record lists 23 posted secondary outcomes, while sponsor-reported analyses are not listed for them. These omissions limit evaluation of prespecification and the robustness of secondary findings.

Finally, the primary estimate is imprecise, and the secondary comparisons are based on smaller outcome-specific groups in some cases. The posted results do not establish the planned composite outcome, and the available information does not support broad conclusions across all outcomes named in the study summary.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Treating COVID-19 With Hydroxychloroquine (TEACH) — NCT04369742.
• Registry record last updated 2022-04-22; status last verified 2022-04.
• Results first posted 2022-04-22.
• 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.

22. Record Summary

The registry record describes NCT04369742, Treating COVID-19 With Hydroxychloroquine (TEACH), as a phase 2, randomized, double-blind, parallel-group trial sponsored by NYU Langone Health. Its arms were hydroxychloroquine and placebo, identified as calcium citrate. The study summary describes a planned five-day course and planned enrollment of 626 adults hospitalized with laboratory-confirmed COVID-19. The registry lists enrollment of 128 and a terminated status.

The listed start date is 2020-04-15, and the primary completion date is 2021-04-01. The posted primary result is serious adverse events through day 30: 14.9% in the hydroxychloroquine group and 16.4% in the placebo group. The record also posts selected secondary results for hospital length of stay, QTc prolongation, and 30-day mortality. The reason for termination, the difference between the summary’s planned primary composite and the posted primary endpoint, and key analysis details are not reported in the registry record.