Analysis of seven trials finds that corticosteroids reduce risk of death by 20 per cent in critically ill COVID-19 patients

  • 2 September 2020

Corticosteroids reduce the risk of death among critically ill COVID-19 patients by 20 per cent, an analysis of seven trials published today [2 September] in the Journal of the American Medical Association (JAMA) has found. The results of three of the trials included in the meta-analysis are also published in JAMA today.

Watch JAMA’s video on the research

The study by an international team of scientists, which was co-ordinated by the World Health Organisation (WHO) with analyses led by National Institute for Health Research (NIHR) researchers at the University of Bristol, looked at mortality over a 28-day period after the start of treatment. They found that treatment with one of the three corticosteroids dexamethasone, hydrocortisone or methylprednisolone led to an estimated 20 per cent reduction in the risk of death. This is equivalent to around 68 per cent of patients surviving after treatment with corticosteroids, compared to around 60 per cent surviving in the absence of corticosteroids. The dexamethasone finding is mainly based on results from the RECOVERY trial, which were included in the analysis.

The analysis is the result of unprecedented co-operation between study teams, guideline developers and journals in response to the global pandemic. This ground-breaking collaboration saw results shared between research teams and with guideline developers before they were published, and papers reporting individual trials, the meta-analysis and international treatment guidelines published simultaneously today.

The seven randomised control trials (the best type of study to examine the effect of a medical intervention) recruited 1,703 critically ill patients in total, spanning five continents and including some of the countries hardest hit by COVID-19. The analysis included critically ill patients from the RECOVERY trial, which reported its findings in June 2020.

The mortality results were consistent across the seven trials with two types of corticosteroid, dexamethasone and hydrocortisone, giving similar effects. Too few patients were included in trials of methylprednisolone to allow its effect to be estimated with precision.

There was evidence of benefit from corticosteroids regardless of whether patients were receiving invasive mechanical ventilation at the time they started treatment. The benefit appeared greater among patients who were not so sick that they needed medicine to support their blood pressure, although the results were not definitive in this regard. The effect of corticosteroids appeared similar regardless of age, sex or how long patients had been ill.

Jonathan Sterne, Professor of Medical Statistics and Epidemiology, University of Bristol and Deputy Director of the NIHR Bristol Biomedical Research Centre (NIHR Bristol BRC), said:

“Our review is good news in the effort to treat COVID-19, and provides important new information that builds on the findings of the RECOVERY trial. Steroids are a cheap and readily available medication, and our analysis has confirmed that they are effective in reducing deaths amongst the people most severely affected by COVID-19. The results were consistent across the trials and show benefit regardless of age or sex.”

The team behind the review includes the lead researchers for all of the seven trials and scientists from Brazil, Canada, China, France, Spain, the UK and the USA. The work is part of the WHO Rapid Evidence Appraisal for COVID-19 Therapies (REACT) initiative.

John Marshall, Professor of Surgery at the University of Toronto, Senior Scientist in the Li Ka Shing Knowledge Institute, and Co-chair of the WHO Working Group on Clinical Characterization and Management, added:

“Even beyond the clear evidence of benefit for an inexpensive and widely available medication, the process of this work – pooling data across seven trials conducted over a period of only three months – highlights the willingness of researchers around the world to share data in a new research model that can bring reliable evidence rapidly to improve the care of patients with COVID-19.”

Janet Diaz, Lead, Clinical response for COVID-19, WHO Emergency Programme, said:

“WHO is committed to transforming science to policy in order to save lives. The COVID-19 pandemic has challenged us to work faster, but not to sacrifice quality and standards. The WHO Rapid Evidence Appraisal for COVID-19 Therapies (REACT) Working Group demonstrates, how in solidarity, science and public health can come together quickly for a common cause. The milestones of pooling of trial data before publication, using that data to inform clinical guidance development and then simultaneous publication of the evidence, evidence synthesis and guidance is unprecedented. I am privileged to have been part of this tremendous collaboration.”


Association between administration of systemic corticosteroids and mortality among critically ill patients with COVID-19: a meta-analysis’ by Jonathan A C Sterne et al. in JAMA

The WHO Rapid Evidence Appraisal for COVID-19 Therapies (REACT) Working Group included: Jonathan A C Sterne, Srinivas Murthy, Janet Diaz, Arthur S Slutsky, Jesús Villar, Derek C Angus, Djillali Annane, Luciano Cesar Pontes Azevedo, Otavio Berwanger, Du Bin, Alexandre B. Cavalcanti, Pierre-Francois Dequin, Jonathan Emberson, David Fisher, Bruno Giraudeau, Anthony C Gordon, Anders Granholm, Cameron Green, Richard Haynes, Nicholas Heming, Julian P T Higgins, Peter Horby, Peter Jüni, Martin J Landray, Amelie Le Gouge, Marie Leclerc, Wei Shen Lim, Flávia R. Machado, Colin McArthur, Ferhat Meziani, Morten Hylander Møller, Anders Perner, Marie W Petersen, Jelena Savovic, Bruno Tomazini, Viviane C. Veiga, Steve Webb and John C Marshall for the WHO COVID-19 Clinical Management and Characterization Working Group.