Trial Design and Oversight
The RECOVERY trial is an investigator-initiated platform trial to evaluate the effects of potential treatments in patients hospitalized with Covid-19. The trial is being conducted at 176 hospitals in the United Kingdom. (Details are provided in the Supplementary Appendix, available with the full text of this article at NEJM.org.) The investigators were assisted by the National Institute for Health Research Clinical Research Network, and the trial is coordinated by the Nuffield Department of Population Health at the University of Oxford, the trial sponsor. Although patients are no longer being enrolled in the hydroxychloroquine, dexamethasone, and lopinavir–ritonavir groups, the trial continues to study the effects of azithromycin, tocilizumab, convalescent plasma, and REGN-COV2 (a combination of two monoclonal antibodies directed against the SARS-CoV-2 spike protein). Other treatments may be studied in the future. The hydroxychloroquine that was used in this phase of the trial was supplied by the U.K. National Health Service (NHS).
Hospitalized patients were eligible for the trial if they had clinically-suspected or laboratory-confirmed SARS-CoV-2 infection and no medical history that might, in the opinion of the attending clinician, put patients at substantial risk if they were to participate in the trial. Initially, recruitment was limited to patients who were at least 18 years of age, but the age limit was removed as of May 9, 2020.
Written informed consent was obtained from all the patients or from a legal representative if they were too unwell or unable to provide consent. The trial was conducted in accordance with Good Clinical Practice guidelines of the International Conference on Harmonisation and was approved by the U.K. Medicines and Healthcare Products Regulatory Agency (MHRA) and the Cambridge East Research Ethics Committee. The protocol with its statistical analysis plan are available at NEJM.org, with additional information in the Supplementary Appendix and on the trial website at www.recoverytrial.net.
The initial version of the manuscript was drafted by the first and last authors, developed by the writing committee, and approved by all members of the trial steering committee. The funders had no role in the analysis of the data, in the preparation or approval of the manuscript, or in the decision to submit the manuscript for publication. The first and last members of the writing committee vouch for the completeness and accuracy of the data and for the fidelity of the trial to the protocol and statistical analysis plan.
Randomization and Treatment
We collected baseline data using a Web-based case-report form that included demographic data, level of respiratory support, major coexisting illnesses, the suitability of the trial treatment for a particular patient, and treatment availability at the trial site. Using a Web-based unstratified randomization method with the concealment of trial group, we assigned patients to receive either the usual standard of care or the usual standard of care plus hydroxychloroquine or one of the other available treatments that were being evaluated. The number of patients who were assigned to receive usual care was twice the number who were assigned to any of the active treatments for which the patient was eligible (e.g., 2:1 ratio in favor of usual care if the patient was eligible for only one active treatment group, 2:1:1 if the patient was eligible for two active treatments, etc.).
For some patients, hydroxychloroquine was unavailable at the hospital at the time of enrollment or was considered by the managing physician to be either definitely indicated or definitely contraindicated. Patients with a known prolonged corrected QT interval on electrocardiography were ineligible to receive hydroxychloroquine. (Coadministration with medications that prolong the QT interval was not an absolute contraindication, but attending clinicians were advised to check the QT interval by performing electrocardiography.) These patients were excluded from entry in the randomized comparison between hydroxychloroquine and usual care.
In the hydroxychloroquine group, patients received hydroxychloroquine sulfate (in the form of a 200-mg tablet containing a 155-mg base equivalent) in a loading dose of four tablets (total dose, 800 mg) at baseline and at 6 hours, which was followed by two tablets (total dose, 400 mg) starting at 12 hours after the initial dose and then every 12 hours for the next 9 days or until discharge, whichever occurred earlier (see the Supplementary Appendix).15 The assigned treatment was prescribed by the attending clinician. The patients and local trial staff members were aware of the assigned trial groups.
A single online follow-up form was to be completed by the local trial staff members when each trial patient was discharged, at 28 days after randomization, or at the time of death, whichever occurred first. Information was recorded regarding the adherence to the assigned treatment, receipt of other treatments for Covid-19, duration of admission, receipt of respiratory support (with duration and type), receipt of renal dialysis or hemofiltration, and vital status (including cause of death). Starting on May 12, 2020, extra information was recorded on the occurrence of new major cardiac arrhythmia. In addition, we obtained routine health care and registry data that included information on vital status (with date and cause of death) and discharge from the hospital.
The primary outcome was all-cause mortality within 28 days after randomization; further analyses were specified at 6 months. Secondary outcomes were the time until discharge from the hospital and a composite of the initiation of invasive mechanical ventilation including extracorporeal membrane oxygenation or death among patients who were not receiving invasive mechanical ventilation at the time of randomization. Decisions to initiate invasive mechanical ventilation were made by the attending clinicians, who were informed by guidance from NHS England and the National Institute for Health and Care Excellence. Subsidiary clinical outcomes included cause-specific mortality (which was recorded in all patients) and major cardiac arrhythmia (which was recorded in a subgroup of patients). All information presented in this report is based on a data cutoff of September 21, 2020. Information regarding the primary outcome is complete for all the trial patients.
For the primary outcome of 28-day mortality, we used the log-rank observed-minus-expected statistic and its variance both to test the null hypothesis of equal survival curves and to calculate the one-step estimate of the average mortality rate ratio in the comparison between the hydroxychloroquine group and the usual-care group. Kaplan–Meier survival curves were constructed to show cumulative mortality over the 28-day period. The same methods were used to analyze the time until hospital discharge, with censoring of data on day 29 for patients who had died in the hospital. We used the Kaplan–Meier estimates to calculate the median time until hospital discharge. For the prespecified composite secondary outcome of invasive mechanical ventilation or death within 28 days (among patients who had not been receiving invasive mechanical ventilation at randomization), the precise date of the initiation of invasive mechanical ventilation was not available, so the risk ratio was estimated instead. Estimates of the between-group difference in absolute risk were also calculated.
All the analyses were performed according to the intention-to-treat principle. Prespecified analyses of the primary outcome were performed in six subgroups, as defined by characteristics at randomization: age, sex, race, level of respiratory support, days since symptom onset, and predicted 28-day risk of death. (Details are provided in the Supplementary Appendix.)
Estimates of rate and risk ratios are shown with 95% confidence intervals without adjustment for multiple testing. The P value for the assessment of the primary outcome is two-sided. The full database is held by the trial team, which collected the data from the trial sites and performed the analyses, at the Nuffield Department of Population Health at the University of Oxford.
The independent data monitoring committee was asked to review unblinded analyses of the trial data and any other information that was considered to be relevant at intervals of approximately 2 weeks. The committee was then charged with determining whether the randomized comparisons in the trial provided evidence with respect to mortality that was strong enough (with a range of uncertainty around the results that was narrow enough) to affect national and global treatment strategies. In such a circumstance, the committee would inform the members of the trial steering committee, who would make the results available to the public and amend the trial accordingly. Unless that happened, the steering committee, investigators, and all others involved in the trial would remain unaware of the interim results until 28 days after the last patient had been randomly assigned to a particular treatment group.
On June 4, 2020, in response to a request from the MHRA, the independent data monitoring committee conducted a review of the data and recommended that the chief investigators review the unblinded data for the hydroxychloroquine group. The chief investigators and steering committee members concluded that the data showed no beneficial effect of hydroxychloroquine in patients hospitalized with Covid-19. Therefore, the enrollment of patients in the hydroxychloroquine group was closed on June 5, 2020, and the preliminary result for the primary outcome was made public. Investigators were advised that any patients who were receiving hydroxychloroquine as part of the trial should discontinue the treatment.