Hydroxychloroquine vs. Azithromycin for Hospitalized Patients With Suspected or Confirmed COVID-19: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Hydroxychloroquine vs. Azithromycin for Hospitalized Patients With Suspected or Confirmed COVID-19 (NCT04329832), a randomized phase 2 study. The primary outcome is COVID Ordinal Outcomes Scale at 14 Days.

2. Clinical Question

3. Trial Design

The registry describes HAHPS as a prospective pragmatic trial comparing hydroxychloroquine with azithromycin among hospitalized patients with suspected or confirmed COVID-19. It is listed as a phase 2, randomized, parallel-group study with no masking. The trial status is completed. The record gives an enrollment of 85; the primary outcome summaries list 42 participants in the hydroxychloroquine arm and 43 in the azithromycin arm.

Random allocation means assignment to the treatment groups was intended to occur by chance. In general, randomization can help balance measured and unmeasured characteristics across groups, although chance imbalances can still occur, particularly in a trial of this size. No details on the sequence-generation method or allocation concealment are reported in the registry record. The open-label designation means the study is listed as having no masking; the record does not specify how knowledge of treatment assignment may have affected outcome assessment or care.

The primary outcome was assessed once on day 14 after enrollment, with enrollment defined as day 0. The registry describes the measure as the COVID-19 WHO ordinal outcomes score, summarized by median and interquartile range. Secondary outcomes listed are hospital-free days at 28 days, ventilator-free days at 28 days, ICU-free days at 28 days, and time to a one-point decrease in the WHO ordinal recovery score. The posted record does not include numerical results for those secondary outcomes.

The registry gives a trial start date of 2020-03-30 and a primary completion date of 2020-07-01. It does not report the study protocol, detailed enrollment criteria, treatment regimens, or a fuller account of trial conduct.

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

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 COVID Ordinal Outcomes Scale at 14 Days, measured over Assessed once on day 14 after enrollment (enrollment is day 0).

The primary endpoint is the COVID-19 WHO ordinal outcomes score assessed on day 14 after enrollment. The registry defines a scale from 0 to 8. Its ordered categories range from no clinical or virological evidence of infection, through activity limitation, hospitalization and increasing levels of respiratory support, to death. A lower score therefore corresponds to a less severe category on the scale, while higher scores correspond to more severe categories.

The score is used to represent clinical status along an ordered range rather than to measure a single symptom or laboratory result. This can capture differences in overall severity, including transitions between levels of care and respiratory support. The registry reports the endpoint as a median with an interquartile range, a suitable descriptive format for an ordered score when the distribution may not be well represented by a mean and standard deviation.

The assessment is specified at a particular follow-up time: once on day 14, with enrollment as day 0. The posted arm medians are 2.0 for both treatments. The interquartile ranges are 2.0 to 3.75 for hydroxychloroquine and 1.5 to 3.0 for azithromycin. The record does not report a statistical comparison of the arm distributions or a confidence interval for a between-arm effect. The endpoint description does not explain how a score was handled if an assessment was unavailable.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a median and interquartile range describe?

For an ordered outcome, the median is the central value after observations are ranked. The interquartile range gives the interval from the lower to the upper quartile, describing the middle half of observations. These summaries are often useful when a measure is ordinal or its distribution is not well described by a symmetric model. They preserve the ordering of the WHO score without requiring the category steps to be treated as equal distances.

For this endpoint, the posted medians are 2.0 in both arms. The interquartile ranges are 2.0 to 3.75 for hydroxychloroquine and 1.5 to 3.0 for azithromycin. The summaries do not reveal how many participants fell into each score category or whether the two distributions overlap substantially. They also do not constitute an estimate of the treatment effect. A formal comparison would need a stated estimand, an analysis method, and an uncertainty measure.

Why is an uncertainty interval useful?

General explanation: a confidence interval expresses the uncertainty around an estimated quantity under a specified statistical method. For a treatment comparison, an interval would indicate which effect sizes are reasonably compatible with the observed data and model assumptions. Its interpretation depends on how the effect is defined and how the data are analyzed. The registry record does not report a confidence interval for the primary between-arm comparison, so the precision of such a contrast cannot be evaluated from the posted summaries alone.

A p-value, if reported, would address the compatibility of data with a specified null hypothesis under a statistical model. It would not measure the size or importance of a treatment effect, nor would it show the probability that the null hypothesis is true. No p-value for the primary comparison is reported in the registry record. The equal medians should therefore be read as descriptive summaries, not as a formal finding of equivalence or absence of a difference.

What does randomization contribute?

General explanation: random assignment can reduce systematic differences in baseline characteristics between treatment groups and supports a causal comparison when the trial is conducted and analyzed appropriately. It does not guarantee that the groups will be identical, and it does not resolve problems caused by missing outcomes, departures from assigned treatment, or selective analysis. The record identifies the trial as randomized, but does not report the randomization procedure, allocation concealment, or analysis population.

The study is listed as parallel-group and unmasked. In a parallel design, participants are assigned to one treatment group rather than receiving both treatments in sequence. In an unmasked study, treatment assignment is not concealed from participants or study personnel as described by the registry. Knowledge of assignment can matter for care decisions, reporting, and outcome assessment; the record does not provide further details about these processes.

How should the planning assumptions be read?

The registry includes planning assumptions for an illustration based on the observed effect: a two-sided significance level of 0.05, equal allocation between arms, and a normal approximation. The stated effect to detect is a fraction of the observed effect. Such calculations describe hypothetical planning scenarios under assumptions; they are not evidence that the trial had a particular probability of detecting an effect, and they are not observed power.

The assumptions alone do not show the resulting participant requirements for any stated effect fraction. No calculated planning values are included in the record details. These planning illustrations should not be mistaken for the trial's own design justification, which is not reported in the registry record. The record also does not report whether an original sample-size calculation was performed or what effect it was intended to detect.

What can be inferred from the posted analysis status?

The registry reports 0 posted analyses. It gives endpoint-level arm summaries for the primary outcome, but no modeled or unadjusted between-arm estimate is included. An arm-by-arm summary and an analysis of a treatment contrast answer different questions: the first describes each group, while the second estimates how outcomes compare and typically includes a measure of uncertainty.

The record marks values as missing or not estimable for a two-arm comparison for another endpoint. It does not provide the analysis method for that endpoint or numerical comparative results. A method cannot be reconstructed from the listed summaries alone. No adjusted analysis, subgroup analysis, interim analysis, or prespecified statistical plan is reported in the registry record.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists one primary outcome and four secondary outcomes: hospital-free days at 28 days, ventilator-free days at 28 days, ICU-free days at 28 days, and time to a one-point decrease in the WHO ordinal recovery score. Multiple outcomes can create a multiplicity issue because examining several hypotheses can increase the chance of at least one apparently unusual result under a collection of null hypotheses. General statistical practice may address this through prespecification or adjustment, but the appropriate approach depends on the trial's aims and analysis plan.

The registry record does not report a multiplicity strategy, such as a hierarchy of tests or an adjustment to significance thresholds. It also does not post numerical results for the listed secondary outcomes. Consequently, neither the findings for those endpoints nor the handling of multiple comparisons can be assessed from the posted record.

Missing-data handling and analysis populations are not reported in the registry record. The primary endpoint summaries have arm counts, but the record does not define whether those counts refer to all randomized participants, participants treated, or participants with available day 14 scores. It does not say whether missing outcomes were imputed, excluded, or handled by another method. Nor does it describe withdrawals, protocol deviations, or adherence to assigned treatment.

These omissions matter because the interpretation of a randomized comparison depends not only on group assignment but also on how outcomes and participants enter the analysis. Without a stated analysis population, missing-data procedure, and multiplicity plan, readers cannot determine whether the posted summaries correspond to the intended primary analysis or how secondary outcome testing was controlled.

11. 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 registry lists serious adverse events for 0 participants in the hydroxychloroquine arm and 0 in the azithromycin arm. It lists other adverse events for 3 participants in the hydroxychloroquine arm and 2 in the azithromycin arm. The deaths field lists 6 for hydroxychloroquine and 1 for azithromycin. These are the posted counts, not estimates of comparative risk.

The record does not describe the criteria used to classify an event as serious or other, the time period covered by the counts, or whether events were assessed as related to treatment. It also does not provide participant-level event descriptions or an analysis comparing the arms. A count of zero listed serious events should not be read as proof that no serious events occurred outside the registry's reporting scope, and the death counts should not be attributed to a drug on the basis of these figures alone.

Safety counts are best interpreted alongside definitions, exposure information, follow-up, and clinical context. Those details are not reported in the registry record. The figures therefore describe what the registry lists, while leaving the comparative safety interpretation uncertain.

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

The posted primary outcome summaries show a median WHO COVID ordinal score of 2.0 at day 14 in each treatment arm. The interquartile ranges differ: 2.0 to 3.75 for hydroxychloroquine and 1.5 to 3.0 for azithromycin. Because the score orders clinical categories from no evidence of infection to death, these summaries describe the central location and spread of observed status on that scale.

The summaries do not establish that the treatments produce the same outcomes, nor do they establish that one treatment produces better outcomes. Equal medians do not imply identical distributions, and different interquartile ranges are not, by themselves, evidence of a statistically reliable treatment difference. The registry does not report a formal between-arm estimate, confidence interval, or p-value for the primary outcome. As a result, the magnitude and precision of a treatment contrast cannot be determined from the posted results.

The record also does not provide numerical findings for the four listed secondary outcomes. It does not report a statistical analysis plan, missing-data method, or clearly defined analysis population. These gaps limit assessment of how complete the outcome summaries are and whether the analysis followed a prespecified approach.

The safety listings provide counts of serious adverse events, other adverse events, and deaths by arm, but do not establish causation or comparative safety. Event definitions and attribution are not reported. The registry record is therefore informative about the design and the descriptive primary outcome summaries, but it is not a complete statistical account of the trial's comparative findings.

In general, a trial result is interpreted by considering the effect estimate, its uncertainty, the design, and the completeness of outcome reporting together. Here, the arm-level median and interquartile range are available, while the central inferential quantities and key analysis details are not reported in the registry record. Conclusions should remain limited to the descriptive information posted.

13. Why This Trial Matters Statistically

This trial illustrates an important distinction in clinical-trial reporting: arm summaries are not the same as a comparative treatment estimate. The registry gives a median and interquartile range for each arm on an ordered clinical scale. Those statistics help describe the observed data, but they do not quantify the uncertainty around a difference between arms. Without a comparison estimate and its precision, readers cannot judge how compatible the results are with a range of possible treatment effects.

The outcome itself is clinically layered. Its ordered categories reflect a spectrum from no evidence of infection to death, including activity limitations, hospitalization, oxygen support, ventilation, and additional organ support. That breadth can make an ordinal endpoint useful for summarizing clinical status, while also making the choice of analysis important. Methods for ordinal outcomes can target different summaries or probabilities, and the registry record does not identify a comparative method for this endpoint.

The record also shows why completeness of reporting matters. It identifies a randomized, unmasked design and lists secondary outcomes, but posts no analyses and no numerical results for those secondary measures. Details about missing data, analysis populations, multiplicity, and the statistical plan are not reported in the registry record. Without these details, readers cannot fully assess the connection between trial design and the results that are posted.

For biostatistical interpretation, this is a case where restraint is essential. The median summaries should be described as observed arm-level data, not transformed into a conclusion about equivalence, benefit, or harm. Similarly, adverse-event and death counts need definitions and context before they can support a comparative safety assessment. The trial matters statistically because it demonstrates both the value of a clear ordered endpoint and the limits of descriptive results when inferential reporting is absent.

14. 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 for interpretation is that the registry does not report a statistical comparison of the primary outcome. The posted medians and interquartile ranges describe each arm, but there is no comparative effect estimate, confidence interval, or p-value in the record. This prevents an assessment of the size and precision of a between-arm difference.

The record does not describe how the day 14 outcome was handled when unavailable, or whether the reported summaries correspond to all randomized participants or another analysis population. Although arm counts are reported with the primary endpoint, their relationship to outcome completeness is not specified. Missing-data procedures, participant flow, treatment adherence, and protocol deviations are not reported in the registry record.

The study is listed as unmasked, but the record gives no further account of outcome assessment or care processes. It also does not report detailed eligibility criteria, dosing, allocation concealment, or the randomization procedure. These omissions limit evaluation of trial conduct and the extent to which the posted comparison reflects the intended randomized contrast.

The four secondary outcomes are listed without numerical results, and the record does not state how multiplicity was addressed. The safety entries lack event definitions, timing, and treatment attribution. Finally, the registry does not provide a protocol or statistical analysis plan, or clarify whether any analyses were performed but not posted. These are reporting limitations; they do not establish that the trial did not collect or analyze additional information.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Hydroxychloroquine vs. Azithromycin for Hospitalized Patients With Suspected or Confirmed COVID-19 — NCT04329832.
• Registry record last updated 2022-04-20; status last verified 2022-04.
• Results first posted 2022-04-20.
• 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.

18. Record Summary

HAHPS is listed as a completed, phase 2, prospective pragmatic, randomized, parallel-group trial with no masking. It compared hydroxychloroquine with azithromycin in hospitalized patients with suspected or confirmed COVID-19. The registry reports an enrollment of 85 and gives primary outcome arm counts of 42 for hydroxychloroquine and 43 for azithromycin.

The primary endpoint was the day 14 COVID-19 WHO ordinal outcomes score. The median was 2.0 in each arm; the interquartile range was 2.0 to 3.75 for hydroxychloroquine and 1.5 to 3.0 for azithromycin. Four secondary outcomes are listed, but their numerical results are not reported in the posted record. No statistical analyses are posted.

The safety record lists 0 serious adverse events in each arm, other adverse events of 3 and 2, and deaths of 6 and 1, respectively. Event definitions and attribution are not reported. The registry record does not provide a comparative effect estimate, confidence interval, p-value, missing-data procedure, or analysis-population definition.