Therapeutics and COVID-19: living guideline, 31 March 2021

© World Health Organization 2020. Some rights reserved. This work is available under the CC BY-NC-SA 3.0 IGO licence. WHO reference number: WHO/2019-nCoV/therapeutics/2020.2 Therapeutics and COVID-19 LIVING GUIDELINE 31 March 2021 Therapeutics and COVID-19 LIVING GUIDELINE 31 March 2021 © World Health Organization 2021. Some rights reserved. This work is available under the CC BY-NC-SA 3.0 IGO licence. WHO reference number: WHO/2019-nCoV/therapeutics/2021.1 3 1. SUMMARY: WHAT IS THIS LIVING GUIDELINE? Clinical question: What is the role of drugs in the treatment of patients with COVID-19? Target audience: The target audience is clinicians and health care decision-makers. Current practice: Current practice to treat COVID-19 is variable, reflecting large-scale uncertainty. Numerous randomized trials of many different drugs are underway to inform practice. This version of the WHO Therapeutics and COVID-19: living guideline contains new information and recommendations on hydroxychloroquine and lopinavir/ritonavir. It follows the preprint publication of results from the WHO SOLIDARITY trial on 15 October 2020 (1) and a peer-reviewed publication on 1 December 2020 (2), which also reported results on remdesivir and interferon- beta. Recommendations: The panel made strong recommendations against the use of hydroxychloroquine and lopinavir/ritonavir in patients with COVID-19, regardless of disease severity. This guidance adds to recommendations published in the previous version with: • a strong recommendation for systemic corticosteroids in patients with severe and critical COVID-19; • a conditional recommendation against systemic corticosteroids in patients with non-severe COVID-19; and • a conditional recommendation against remdesivir in hospitalized patients with COVID-19. How this guideline was created: This living guideline is an innovation from the World Health Organization (WHO), driven by the urgent need for global collaboration to provide trustworthy and evolving COVID-19 guidance informing policy and practice worldwide. WHO has partnered with the non-profit Magic Evidence Ecosystem Foundation (MAGIC) for methodologic support and development and dissemination of living guidance for COVID-19 drug treatments, based on a living systematic review and network analysis (3). An international Guideline Development Group (GDG) of content experts, clinicians, patients, ethicists and methodologists produced recommendations following standards for trustworthy guideline development using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach. No conflict of interest was identified for any panel member. The latest evidence: The recommendation on hydroxychloroquine was informed by results from a systematic review and network meta-analysis (NMA) that pooled data from 30 trials with 10 921 participants with COVID-19 (3). Lopinavir/ritonavir was informed by the same analysis that pooled data from 7 trials with 7429 participants (3). The trials for both drugs included inpatients and outpatients. The resulting GRADE evidence summary suggested that hydroxychloroquine probably does not reduce mortality (odds ratio 1.11, 95% confidence interval: [CI] 0.95–1.31; absolute effect estimate 10 more deaths per 1000 patients, 95% CI: from 5 fewer – 28 more deaths per 1000 patients; moderate certainty evidence) or need for mechanical ventilation. Lopinavir/ritonavir also probably does not reduce mortality (odds ratio 1.00, 95% CI: 0.82–1.20; absolute effect estimate 0 fewer deaths per 1000 patients, 95% CI: from 17 fewer – 19 more deaths per 1000 patients; moderate certainty evidence) or need for mechanical ventilation. Both hydroxychloroquine and lopinavir/ritonavir may the risk of diarrhoea and nausea/vomiting (low certainty evidence). There was no indication of a credible subgroup effect for either intervention based on disease severity or age, and no credible subgroup effect by dose for hydroxychloroquine. Understanding the recommendations: When moving from the evidence to the strong recommendation against the use of hydroxychloroquine and lopinavir/ritonavir in patients with COVID-19, the panel emphasized the evidence suggesting no reduction in mortality, need for mechanical ventilation, and other patient-important outcomes. There were also potential for harms with both drugs including diarrhoea, nausea/vomiting, and other adverse effects that were not elucidated in the available trials. The panel did not anticipate important variability when it comes to patient values and preferences. In addition, the panel decided that contextual factors such as resources, feasibility, acceptability and equity for countries and health care systems were unlikely to alter the recommendation. 4 Info box This WHO Therapeutics and COVID-19: living guidelines now includes strong recommendations against the use of hydroxychloroquine and lopinavir/ritonavir. This update was initiated after publication of the WHO SOLIDARITY trial (1,2). Please view Section 1 for an executive summary of the guidance. The first version of the living WHO guideline published 2 September 2020 provides recommendations for corticosteroids; the second version published 20 November 2020 provides recommendations for remdesivir, with no changes for either of these drugs made as part of this update. This is a living guideline, so the recommendations included here will be updated, and new recommendations will be added on other therapies for COVID-19. The guideline is therefore written, disseminated and updated here in MAGICapp, with a format and structure aiming to make it user-friendly and easy to navigate while accommodating for dynamically updated evidence and recommendations, focusing on what is new, while keeping existing recommendations within the guideline. Please visit the WHO website for the latest version of the guidance, also available in the BMJ as Rapid Recommendations together with the living network meta-analysis (NMA), a major evidence source for the guidelines (3). The updated living NMA informing the recommendation on both hydroxychloroquine and lopinavir/ritonavir has been published in the BMJ (3). The same team performed a systematic review and meta-analysis on adverse effects from these drugs. This paper is currently available as preprint through MedRxiv. 2. ABBREVIATIONS ARDS acute respiratory distress syndrome CAP community-acquired pneumonia CI confidence interval GDG guideline development group GRADE Grading of Recommendations Assessment, Development and Evaluation HIV human immunodeficiency virus MAGIC Magic Evidence Ecosystem Foundation NMA network meta-analysis PICO population, intervention, comparator, outcome PMA prospective meta-analysis RCT randomized controlled trial SAE serious adverse event WHO World Health Organization 3. BACKGROUND As of 14 December 2020, over 70 million people worldwide have been diagnosed with COVID-19, according to the WHO dashboard (4). The pandemic has so far claimed more than 1.6 million lives, and many areas of the world are experiencing a resurgence in cases. The COVID-19 pandemic – and the explosion of both research and misinformation – has highlighted the need for trustworthy, accessible and regularly updated (living) guidance to place emerging findings into context and provide clear recommendations for clinical practice (5). This living guideline responds to emerging evidence from randomized controlled trials (RCTs) on existing and new drug treatments for COVID-19. More than 2800 trials investigating interventions for COVID-19 have been registered or are ongoing (see section on emerging evidence) (6). Among these are large national and international platform trials (e.g. RECOVERY, WHO SOLIDARITY and DISCOVERY) that recruit very large numbers of patients in many countries, with a pragmatic and adaptive design (2,7). These platform trials are currently investigating and reporting on drugs such as antiviral monoclonal antibodies and immunomodulators. This rapidly evolving evidence landscape requires trustworthy interpretation and expeditious clinical practice guidelines to inform clinicians, patients, governments, ministries and health administrators. 5 3.1 What triggered this version of the guideline? This third version of the WHO living guideline addresses the use of hydroxychloroquine and lopinavir/ritonavir in patients with COVID-19. It follows the preprint publication and peer-reviewed publication of the WHO SOLIDARITY trial on 15 October 2020, reporting results on treatment with remdesivir, hydroxychloroquine and lopinavir/ritonavir in hospitalized patients with COVID-19 (1,2). The role of these drugs in clinical practice has remained uncertain, with limited prior trial evidence. The WHO SOLIDARITY trial adds 11 266 randomized patients (2570 to remdesivir, 954 to hydroxychloroquine, and 1411 to lopinavir/ritonavir, 6331 to usual care) and holds the potential to change practice (2). In response to the release of SOLIDARITY data, the WHO GDG started with developing trustworthy recommendations on remdesivir (published 20 November 2020), and now provides recommendations on hydroxychloroquine and lopinavir/ritonavir. Hydroxychloroquine and chloroquine are anti-inflammatory agents that work through blocking of Toll-like receptors reducing dendritic cell activation. Hydroxychloroquine is used to treat rheumatoid arthritis and systemic lupus erythematosus. Chloroquine is listed in the WHO Model List of Essential Medicines as an antimalarial, for use for the treatment of P. vivax infection. Chloroquine has an antiviral effect against many viruses in vitro, including SARS-CoV-2, but a clinically useful antiviral effect has not been shown for any viral infection. Lopinavir is a protease-inhibitor antiretroviral agent, commonly used in combination with ritonavir which increases the serum concentration of lopinavir. This combination drug is used to treat and prevent human immunodeficiency virus (HIV) infection. 3.2 Who made this guideline? As detailed in Section 4. Methods, the WHO convened a standing GDG with 28 clinical content experts, 4 patient- partners and one ethicist, headed by a clinical chair (Dr Michael Jacobs) and two methods chairs (Dr Reed Siemieniuk [hydroxychloroquine] and Dr Bram Rochwerg [lopinavir/ritonavir]). WHO selected GDG members to ensure global geographical representation, gender balance, and appropriate technical and clinical expertise. No panel member had a conflict of interest. The MAGIC Evidence Ecosystem Foundation (MAGIC) provided methodological experts with high-level expertise in standards and methods for systematic reviews and guideline development, including GRADE; in addition, MAGIC offered innovations in processes (BMJ Rapid Recommendations) and platforms (MAGICapp) for developing living guidance in user-friendly formats. The WHO has in place Agreements for Performance of Work with this entity for these two deliverables. The methodological experts were not involved in the formulation of recommendations. MAGIC also worked with the BMJ to coordinate the simultaneous scientific publication of the living WHO guidelines (8). 3.3 How to use this guideline This is a living guideline from the WHO. Recommendations will be updated, and new recommendations will be added on other therapies for COVID-19 (8). The guideline is written, disseminated and updated in MAGICapp, with a format and structure aiming to make it user-friendly and easy to navigate (9). It accommodates dynamic updating of evidence and recommendations that can focus on what is new while keeping existing recommendations, as appropriate, within the guideline. Section 4 outlines key methodological aspects of the living guideline process. The guideline is available here in MAGICapp in online, multilayered formats and via: • WHO website in PDF format • WHO Academy app • BMJ Rapid Recommendations (8). The purpose of the MAGICapp online formats and additional tools, such as the infographics made by the BMJ, is to make it easier to navigate and use the guideline in busy clinical practice. The online multilayered formats are designed to allow end-users to find recommendations first and then drill down to find supporting evidence and other information pertinent to applying the recommendations in practice, including tools for shared decision-making (clinical encounter decision aids). 6 4. METHODS: HOW THIS GUIDELINE WAS CREATED The living WHO guideline is developed according to standards and methods for trustworthy guidelines, making use of an innovative process to achieve efficiency in dynamic updating of recommendations. The methods are aligned with the WHO handbook for guideline development and according to a pre-approved protocol (planning proposal) by the Guideline Review Committee. Related guidelines This living WHO guideline for COVID-19 treatments will be related to the larger, more comprehensive guidance for Clinical management of COVID-19: interim guidance, which has a wider scope of content and is currently being updated and will also become available on the MAGICapp (9). The first two WHO living guidelines, addressing corticosteroids and remdesivir, were disseminated via the WHO website, BMJ and MAGICapp. Timing This guidance aims to be trustworthy and living; dynamically updated and globally disseminated once new evidence warrants a change in recommendations for COVID-19 therapeutics (10). We aim for an ambitious timeframe from trials that trigger the guideline development process to WHO publication within 1 month, while maintaining standards and methods for trustworthy guidelines (WHO handbook of guideline development). Stepwise approach Here we outline the stepwise approach we take to improve efficiency and timeliness of the living, trustworthy guidance, in the development and dissemination of the recommendations. To do so, various processes occurred simultaneously. Step 1: Evidence monitoring and mapping and triggering of evidence synthesis Comprehensive daily monitoring of all emerging RCTs occurs on a continuous basis, within the context of the living systematic review and NMA, using experienced information specialists, who look at all relevant information sources for new RCTs addressing interventions for COVID-19. Once practice-changing evidence is identified, such as in this case, the SOLIDARITY trial preprint, the WHO Therapeutics Steering Committee triggered the guideline development process. With the Guidance Support Collaboration Committee (see Acknowledgements), PICO (population, intervention, comparator, outcome) development and construction of evidence summaries addressing the intervention of interest are initiated. The trigger for producing or updating specific recommendations is based on the following: • likelihood to change practice; • sufficient RCT data on therapeutics to inform the high-quality evidence synthesis living systematic review; • relevance to a global audience. Step 2: Convening the GDG The pre-selected expert panel (see Acknowledgments) convened on five occasions. The first meeting, held 13 October 2020, reviewed the basics of GRADE methodology; including formulating PICO questions and subgroups of interests, assessment of certainty of evidence, incorporating patients’ values and preferences, and prioritization of patient-important outcomes. The second meeting, held on 20 October 2020, finalized the outcome prioritization, PICOs and pre-specified subgroups for this specific question. At the third meeting, held on 23 October 2020, a Q&A session was held with the individual study investigators and biostatisticians: SOLIDARITY (Drs Ana Maria Henao Restrepo and Richard Peto), ACTT-1 (Drs Lori Dodd and John Beigel) and RECOVERY (Drs Peter Horby and Jonathan Emberson). These first three meetings addressed issues related to remdesivir, hydroxychloroquine and lopinavir/ritonavir. At the fourth meeting, held on 17 November 2020, evidence summaries were shown to the GDG panel, including pre-specified subgroup analysis, and a recommendation for lopinavir/ritonavir was drafted. At the fifth meeting, on 24 November 2020, evidence summaries including subgroup analysis for hydroxychloroquine were shown, and a recommendation addressing this intervention was drafted. Step 3: Evidence synthesis The living systematic review/NMA team, as requested by the WHO Therapeutics Steering Committee and coordinated by the Guidance Support Collaboration Committee, was ready to perform an independent systematic review to examine the benefits and harms of the intervention. The systematic review team is multidisciplinary and made up of systematic review experts, clinical experts, clinical epidemiologists, graduate students and biostatisticians. The team 7 has expertise in GRADE methodology and rating certainty of evidence specifically in NMAs. The NMA team was informed of the deliberations from the initial two GDG meetings in order to guide the NMA, specifically focusing on the outcomes and subgroups prioritized by the panel. To conduct the subgroup analysis of high vs low dose of hydroxychloroquine, Professor Andrew Owen (see Acknowledgments) was engaged to provide direction on how to analyse different dosing regimens of hydroxychloroquine. Based on pharmacokinetic data of the different dosing regimens, Professor Owen and the methods support team recommended analysing cumulative dose as a continuous variable, with a sensitivity analysis using predicted serum trough concentration on Day 3 (a measure of early dosing) for efficacy outcomes. Step 4: Final recommendations The GDG panel members are responsible for the following critical activities: • reviewing the evidence synthesis and summary of finding tables (presented by the NMA team) and from the evidence-drafting recommendations; • advising on the priority questions and scope of guidance; • advising on the choice of important outcomes for decision-making; • commenting on the evidence used to inform the guideline; • advising on the interpretation of the evidence, with explicit consideration of the overall balance of risks and benefits; • formulating recommendations, taking into account diverse values and preferences according to GRADE. The GRADE approach provided the framework for establishing evidence certainty and generating both the direction and strength of recommendations (11). Although a priori voting procedures were established at the outset, in case consensus was not reached, these procedures were not necessary for this recommendation, which reached consensus amongst the panel. The following key factors were used to formulate transparent and trustworthy recommendations: • absolute benefits and harms for all patient-important outcomes through structured evidence summaries (e.g. GRADE summary of findings tables) (12); • quality/certainty of the evidence (13); • values and preferences of patients (14); • resources and other considerations (including considerations of feasibility, applicability, equity) (14); • each outcome will have an effect estimate and CI, with a measure of certainty in the evidence, as presented in summary of findings tables. If such data are not available, narrative summaries will be provided; • recommendations will be rated as either conditional or strong, as defined by GRADE. If the panel members disagree regarding the evidence assessment or strength of recommendations, WHO will apply voting according to established methods. Step 5: External and internal review The WHO guideline was then reviewed by pre-specified external reviewers (see Acknowledgements) and then approved by the WHO Publication Review Committee. 5. THE LATEST EVIDENCE This section outlines the information the GDG panel requested and used in making their recommendations for hydroxychloroquine and lopinavir/ritonavir. Benefits and harms The GDG panel requested an update of the living NMA of RCTs of drug treatments for COVID-19, based around important clinical questions to be addressed in the recommendations. The GDG members prioritized outcomes (rating from 1 [not important] to 9 [critical]) taking a patient perspective (Table 1). The panel’s questions were structured using the PICO format (see evidence profile under the recommendations). 8 Table 1. Panel outcome rating from a patient perspective Note: 1: not important, 9: critically important. For hydroxychloroquine: The evidence summary was based on 30 trials and 10 921 participants for which the NMA provided relative estimates of effect for patient-important outcomes (Table 2). Five of the trials (414 total participants) randomized some patients to chloroquine. For lopinavir/ritonavir: The evidence summary was based on 7 trials with 7429 participants (Table 3). Of note, none of the included studies enrolled children or adolescents under the age of 19 years old. Table 2. Summary of trials and trial characteristics informing the hydroxychloroquine recommendation (trials = 30, total patients = 10 921) Geographic region Region of the Americas South-East Asia Region Western Pacific Region European Region Eastern Mediterranean Region Region of the Americas (12 trials, 2358 patients) South-East Asia and Western Pacific Regions (7 trials, 731 patients) European Region (10 trials, 7638 patients) Eastern Mediterranean Region (1 trial, 194 patients) Severity of illnessa Non-severe Severe Critically ill Mild/Moderate (10 trials, 2436 patients) Severe (1 trials, 479 patients) Critically ill (0 trials, 0 patients) Mechanically ventilated at baselineb Mean (range), % 3.23 (0–16.8) Agec Mean (range of means), years 50.8 (32.9–77.0) Sexd Mean (range of means), % women 46.9 (30.0–71.0) Loading doses Day 1e Mean (range of means), mg 1010 (800–1600) Total cumulative dosesf Median (range), mg 4000 (2000–11200) Duration of therapyg Median (range), days 7 (4–16) Type of care n (%) inpatient n (%) outpatient Inpatient: 9549 (87.4) Outpatient: 1372 (12.6) Trial participants Median (range) 364 (2–4716) Concomitant use of corticosteroidsh Mean (range across trials that report this), % 12.61 (8.0–19.5) Notes: a 19 trials did not report the disease severity of patients. b 19 trials did not report the proportion of mechanical ventilation at baseline. c Based on 15 trials and 8006 patients. For the other 15 trials: 1 trial did not report the age of patients; and the other 14 trials reported that the age of patients were ≥ 12, 18 or 40. d 14 trials did not report the sex of patients. e 10 trials did not use a loading dose. f 1 trial reported range of treatment duration. g 1 trial reported range of treatment duration. h 23 trials did not report the concomitant use of corticosteroids. 9 Table 3. Summary of trials and trial characteristics informing the lopinavir/ritonavir recommendation (trials = 7, total patients = 7429) Geographic region Region of the Americas South-East Asia Region Western Pacific Region European Region Eastern Mediterranean Region Region of the Americas (0 trials, 0 patients) South-East Asia and Western Pacific Regions (5 trials, 535 patients) European Region (2 trials, 6894 patients) Middle East (0 trials, 0 patients) Severity of illnessa Non-severe Severe Critically ill Mild/Moderate (4 trials, 336 patients) Severe (1 trials, 199 patients) Critically ill (0 trials, 0 patients) Mechanically ventilated at baselineb Mean (range), % 7.3 (0–16.1) Agec Mean (range of means), years 52.6 (42.5–66.2) Sex Mean (range of means), % women 48.7 (38.9–61.7) Loading doses Day 1d Mean (range of means), mg Total cumulative doses (lopinavir/ ritonavir)e Median (range), mg 11200/2800(8000–11 200/2000–2800) Duration of therapyf Median (range), days 14 (10–14) Type of care n (%) inpatient n (%) outpatient Inpatient: 7429 (100) Outpatient: 0 (0) Trial participants Median (range) 101 (60–5040) Concomitant use of corticosteroidsg Mean (range across trials that report this), % 17.1 (0–32.3) Notes: a 2 trials did not report the disease severity of patients. b 3 trials did not report proportion of mechanical ventilation at baseline. c 2 trials did not report the age of patients. d No trial reported loading dose. e 1 trial did not report cumulative doses; 2 trials only reported range of treatment duration. f 1 trial did not report the duration of therapy, 2 trials used a range of treatment duration. g 2 trials did not report the concomitant use of corticosteroids. Subgroup analysis For both hydroxychloroquine and lopinavir/ritonavir, the GDG panel requested subgroup analyses based on age (considering children vs younger adults [e.g. under 70 years] vs older adults [e.g. 70 years or older]), and illness severity (non-severe vs severe vs critical COVID – see subgroup under Section 7.1 Hydroxychloroquine recommendations for details). The GDG discussed other potential subgroups of interest including time from onset of symptoms until initiation of therapy and concomitant medications, but recognized that these analyses would not be possible without access to individual participant data and/or more detailed reporting from the individual trials. To this last point, the panel recognized that usual care is likely variable between centres, regions and evolved over time. However, given all of the data come from RCTs, use of these co-interventions that comprise usual care should be balanced between study patients randomized to either the intervention or usual care arms. For hydroxychloroquine alone, the panel also requested an analysis based on whether or not it was co-administered with azithromycin. Following the panel’s request, the NMA team performed subgroup analyses in order to assess for effect modification which, if present, could mandate distinct recommendations by subgroups. From the data available from the included trials, subgroup analysis was only possible for severity of illness and age examining the outcome of mortality. This subgroup analysis was performed using a Bayesian analysis which incorporated meta-regression using study as a random effect. The panel used a pre-specified framework incorporating the ICEMAN tool to assess the credibility of subgroup findings (15). The panel also requested a subgroup analysis based on high dose vs low dose hydroxychloroquine. A categorical approach to hydroxychloroquine dosing proved impossible because the trials used varying loading doses, continuation doses and durations. Therefore, in collaboration with a pharmacology expert (Professor Andrew Owen), we modelled the expected serum concentrations over time. We hypothesized that higher trough concentrations early in the treatment course (e.g. trough concentration on Day 3) might be more effective than lower early trough concentrations. We also hypothesized that higher maximum serum concentrations (e.g. peak concentration on the last day) might result in higher risk of adverse effects than lower maximum serum concentrations. In our pharmacokinetic model, the 10 cumulative dose was highly correlated with all measures of serum concentrations on Day 3 and the final day of treatment, and therefore we decided to use cumulative dose as the primary analysis. Day 3 trough concentration was least strongly correlated with total cumulative dose (R2 = 0.376) and therefore we performed a sensitivity subgroup analysis with predicted Day 3 trough concentrations for efficacy outcomes. Baseline risk estimates (prognosis of patients with COVID-19): informing absolute estimates of effect The evidence summaries that informed the guideline recommendation reported the anticipated absolute effects of hydroxychloroquine and lopinavir/ritonavir compared with usual care across all patient-important outcomes, with explicit judgments of certainty in the evidence for each outcome. The absolute effects of treatment are informed by the prognosis (i.e. baseline risk estimates) combined with the relative estimates of effects (e.g. risk ratio, odds ratio) obtained from the NMA. The control arm of the WHO SOLIDARITY trial (2), performed across a wide variety of countries and geographical regions, was identified by the GDG panel as representing the most relevant source of evidence to make the baseline risk estimates for the outcomes of mortality and mechanical ventilation. The rationale for selecting the WHO SOLIDARITY trial was to reflect the overall prognosis of the global population for which the WHO guideline recommendations are made. When applying the evidence to a particular patient or setting, the individual or setting’s risk of mortality and mechanical ventilation should be considered. In view of the study designs, the GDG determined that for other outcomes using the median or mean of all patients randomized to usual care across the included studies would provide the most reliable estimate of baseline risk. Values and preferences There were insufficient published data to provide the GDG with an informative systematic review of studies describing patients’ experiences or values and preferences on treatment decisions for COVID-19 drug treatments. The GDG therefore relied on their own judgments of what well-informed patients would value after carefully balancing the benefits, harms and burdens of treatment and their subsequent treatment preferences. The GDG included four patient-representatives who had lived experience with COVID-19. The GDG agreed that the following values and preferences would be representative of those of typical well-informed patients: • Mortality would be the outcome most important to patients, followed by need and duration of mechanical ventilation, time to clinical improvement, and serious intervention-related adverse events. • Most patients would be reluctant to use a medication for which the evidence left high uncertainty regarding effects on the outcomes listed above. This was particularly so when evidence suggested treatment effects, if they do exist, are small, and the possibility of important harm remains. • In an alternative situation with larger benefits and less uncertainty regarding both benefits and harms, more patients would be inclined to choose the intervention. The GDG acknowledged, however, that values and preferences are likely to vary. There will be patients inclined to use a treatment in which evidence has not excluded important benefit, particularly when the underlying condition is potentially fatal. On the other hand, there will be those who have a high threshold for likely benefit before they will choose the intervention. Although the GDG focused on an individual patient perspective, they also considered a population perspective in which feasibility, acceptability, equity and cost are important considerations. 11 6. WHO DO THE RECOMMENDATIONS APPLY TO? The guideline for COVID-19 therapeutics applies to patients with COVID-19. For some drugs (such as corticosteroids), recommendations may differ based on the severity of COVID-19 disease. The GDG elected to use the WHO severity definitions based on clinical indicators, adapted from the WHO COVID-19 disease severity categorization (see below) (16). These definitions avoid reliance on access to health care to define patient subgroups. WHO severity definitions • Critical COVID-19: Defined by the criteria for acute respiratory distress syndrome (ARDS), sepsis, septic shock or other conditions that would normally require the provision of life-sustaining therapies such as mechanical ventilation (invasive or non-invasive) or vasopressor therapy. • Severe COVID-19: Defined by any of: o oxygen saturation < 90% on room air; o respiratory rate > 30 breaths/min in adults and children > 5 years old; ≥ 60 breaths/min in children < 2 months old; ≥ 50 in children 2–11 months old; and ≥ 40 in children 1–5 years old; o signs of severe respiratory distress (accessory muscle use, inability to complete full sentences, and, in children, very severe chest wall indrawing, grunting, central cyanosis, or presence of any other general danger signs). • Non-severe COVID-19: Defined as absence of any criteria for severe or critical COVID-19. Caution: The panel noted that the oxygen saturation threshold of 90% to define severe COVID-19 was arbitrary and should be interpreted cautiously when used for determining disease severity. For example, clinicians must use their judgment to determine whether a low oxygen saturation is a sign of severity or is normal for a given patient with chronic lung disease. Similarly, a saturation > = 90–94% on room air is abnormal (in patient with normal lungs) and can be an early sign of severe disease, if the patient is on a downward trend. Generally, if there is any doubt, the panel suggested erring on the side of considering the illness as severe. The infographic illustrates these three disease severity groups and key characteristics to apply in practice. Infographic co-produced by BMJ and MAGIC; designer Will Stahl-Timmins (see BMJ Rapid Recommendations). 37 References 1. Pan H, Peto R, Karim Q, Alejandria M, Henao-Restrepo AM, García C et al. Repurposed antiviral drugs for COVID-19: interim WHO SOLIDARITY trial results. MedRxiv. 2020. 2. WHO Solidarity Trial Consortium. Repurposed antiviral drugs for COVID-19: interim WHO Solidarity Trial Results. N Engl J Med. 2020. doi: 10.1056/NEJMoa2023184. 3. Siemieniuk RA, Bartoszko JJ, Ge L, Zeraatkar D, Izcovich A, Kum E et al. Drug treatments for covid-19: living systematic review and network meta-analysis (Update 2). BMJ. 2020;370:m2980. 4. WHO. WHO coronavirus disease (COVID-19) dashboard. Geneva: World Health Organization; 2020. 5. Naci H, Kesselheim AS, Røttingen J-A, Salanti G, Vandvik PO, Cipriani A. Producing and using timely comparative evidence on drugs: lessons from clinical trials for covid-19. BMJ. 2020;371:m3869. 6. Maguire BJ, Guérin PJ. A living systematic review protocol for COVID-19 clinical trial registrations. Wellcome Open Research 2020;5:60. 7. RECOVERY Collaborative Group, Horby P, Lim WS, Emberson JR, Mafham M, Bell JL et al. Dexamethasone in hospitalized patients with Covid-19 – preliminary report. N Engl J Med. 2020. doi: 10.1056/NEJMoa2021436. 8. Lamontagne F, Agoritsas T, Macdonald H, Leo Y-S, Diaz J, Agarwal A et al. A living WHO guideline on drugs for covid-19. BMJ. 2020;370:m3379. 9. Vandvik PO, Brandt L, Alonso-Coello P, Treweek S, Akl EA, Kristiansen A et al. Creating clinical practice guidelines we can trust, use, and share: a new era is imminent. Chest. 2013;144(2):381-389. 10. Qaseem A, Forland F, Macbeth F, Ollenschläger G, Phillips S, van der Wees P. Guidelines International Network: toward international standards for clinical practice guidelines. Ann Int Med. 2012;156(7):525-31. 11. Guyatt GH, Oxman AD, Vist GE, Kunz R, Falck-Ytter Y, Alonso-Coello P et al. GRADE: an emerging consensus on rating quality of evidence and strength of recommendations. BMJ. 2008;336(7650):924-6. 12. Guyatt G, Oxman AD, Akl EA, Kunz R, Vist G, Brozek J et al. GRADE guidelines: 1. Introduction-GRADE evidence profiles and summary of findings tables. J Clin Epidemiol. 2011;64(4):383-94. 13. Balshem H, Helfand M, Schünemann HJ, Oxman AD, Kunz R, Brozek J et al. GRADE guidelines: 3. Rating the quality of evidence. J Clin Epidemiol. 2011;64(4):401-6. 14. Andrews JC, Schünemann HJ, Oxman AD, Pottie K, Meerpohl JJ, Coello PA et al. GRADE guidelines: 15. Going from evidence to recommendation-determinants of a recommendation’s direction and strength. J Clin Epidemiol. 2013;66(7):726-35. 15. Schandelmaier S, Briel M, Varadhan R, Schmid CH, Devasenapathy N, Hayward RA et al. Development of the Instrument to assess the Credibility of Effect Modification Analyses (ICEMAN) in randomized controlled trials and meta-analyses. CMAJ. 2020;192(32):E901-E906. 16. WHO. Clinical management of COVID-19: interim guidance. Geneva: World Health Organization; 2020. 17. NIH. Coronavirus disease 2019 (COVID-19) treatment guidelines. Bethesda (MD): National Institutes of Health. 2020. 18. Rochwerg B, Agarwal A, Zeng L, Leo Y-S, Appiah JA, Agoritsas T et al. Remdesivir for severe covid-19: a clinical practice guideline. BMJ. 2020;370:m2924. 19. Beigel JH, Tomashek KM, Dodd LE, Mehta AK, Zingman BS, Kalil AC et al. Remdesivir for the treatment of Covid-19 – final report. N Engl J Med. 2020;383;1813-1826. 20. Spinner CD, Gottlieb RL, Criner GJ, Arribas López JR, Cattelan AM, Soriano Viladomiu A et al. Effect of remdesivir vs standard care on clinical status at 11 days in patients with moderate COVID-19: a randomized clinical trial. JAMA. 2020;324(11):1048-1057. 38 21. Wang Y, Zhang D, Du G, Du R, Zhao J, Jin Y et al. Remdesivir in adults with severe COVID-19: a randomised, double-blind, placebo-controlled, multicentre trial. Lancet. 2020;395(10236):1569-1578. 22. Persaud N, Jiang M, Shaikh R, Bali A, Oronsaye E, Woods H et al. Comparison of essential medicines lists in 137 countries. Bull World Health Organ. 2019;97(6):394-404C. 23. The WHO Rapid Evidence Appraisal for COVID-19 Therapies (REACT) Working Group, Sterne JAC, Murthy S, Diaz JV, Slutsky AS, Villar J et al. Association between administration of systemic corticosteroids and mortality among critically ill patients with COVID-19: a meta-analysis. JAMA. 2020;324(13):1330-1341. 24. Corral L, Bahamonde A, delas Revillas FA, Gomez-Barquero J, Abadia-Otero J, Garcia-Ibarbia C et al. GLUCOCOVID: a controlled trial of methylprednisolone in adults hospitalized with COVID-19 pneumonia. MedRxiv. 2020. 25. Villar J. Efficacy of dexamethasone treatment for patients with ARDS caused by COVID-19 (DEXA-COVID19). ClinicalTrials.gov [Internet]. Bethesda (MD): National Library of Medicine (US). Identifier NCT04325061 (https://clinicaltrials.gov/ct2/show/NCT04325061, accessed 15 December 2020). 26. Tomazini BM, Maia IS, Cavalcanti AB, Berwanger O, Rosa RG, Veiga VC et al. Effect of dexamethasone on days alive and ventilator-free in patients with moderate or severe acute respiratory distress syndrome and COVID-19: the CoDEX Randomized Clinical Trial. JAMA. 2020;324(13):1307-1316. 27. Dequin P-F, Heming N, Meziani F, Plantefève G, Voiriot G, Badié J et al. Effect of hydrocortisone on 21-day mortality or respiratory support among critically ill patients with COVID-19: a randomized clinical trial. JAMA. 2020;324(13):1298-1306. 28. The Writing Committee for the REMAP-CAP Investigators, Angus DC, Derde L, Al-Beidh F, Annane D, Arabi Y, Beane A et al. Effect of hydrocortisone on mortality and organ support in patients with severe COVID-19: the REMAP-CAP COVID-19 Corticosteroid Domain Randomized Clinical Trial. JAMA 2020;324(13):1317- 1329. 29. Glucocorticoid therapy for COVID-19 critically ill patients with severe acute respiratory failure (Steroids-SARI). ClinicalTrials.gov [Internet]. Bethesda (MD): National Library of Medicine (US). Identifier NCT04244591 (https://www.clinicaltrials.gov/ct2/show/NCT04244591, accessed 15 December 2020). 30. Jeronimo CMP, Farias MEL, Val FFA, Sampaio VS, Alexandre MAA, Melo GC et al. Methylprednisolone as adjunctive therapy for patients hospitalized with COVID-19 (Metcovid): a randomised, double-blind, phase IIb, placebo-controlled trial. Clin Infect Dis. 2020. doi: 10.1093/cid/ciaa1177. 31. WHO. Hospital care for adolescents and adults: guidelines for the management of common illnesses with limited resources. Integrated Management of Adolescent and Adult Illness (IMAI). Geneva: World Health Organization; 2011. 32. Rochwerg B, Oczkowski SJ, Siemieniuk RAC, Agoritsas T, Belley-Cote E, D’Aragon F et al. Corticosteroids in sepsis: an updated systematic review and meta-analysis. Crit Care Med. 2018;46(9):1411-1420. 33. Ye Z, Wang Y, Colunga-Lozano LE, Prasad M, Tangamornsuksan W, Rochwerg B et al. Efficacy and safety of corticosteroids in COVID-19 based on evidence for COVID-19, other coronavirus infections, influenza, community-acquired pneumonia and acute respiratory distress syndrome: a systematic review and meta- analysis. CMAJ. 2020;192(27):E756-E767. 34. WHO. Coronavirus disease (COVID-19): dexamethasone – Q&A. Geneva: World Health Organization; 2020 (https://www.who.int/news-room/q-a-detail/coronavirus-disease-covid-19-dexamethasone, accessed 15 December 2020).