HomeResourcesGeneral resourcesThese steroids can help treat some COVID-19 patients, but not all

These steroids can help treat some COVID-19 patients, but not all

For patients who are very ill with COVID-19, 10 days of treatment with anti-inflammatories can work wonders, but the opposite is true for patients who are less ill. Read the latest guidelines from the World Health Organisation and for South Africa.

Resource details:

Publication title: Corticosteroids for COVID-19
Author(s): The World Health Organisation (WHO)
Publication date: 2 September 2020

Key take-aways from the guidelines:

What the guidelines are about:

These guidelines are designed to help healthcare professionals and policymakers to make evidence-based decisions while treating COVID-19 patients. It’s based on an ongoing analysis of the available evidence of the use of corticosteroids such as dexamethasone to treat COVID-19 patients. Corticosteroids are medicines used to help treat inflamed parts of the body and can help relieve swollen areas, the United States nonprofit the Mayo Clinic explains on its website. 

How were the guidelines put together?  

  • The guideline is a collaboration between the WHO and the research non-profit Magic Evidence Ecosystem Foundation (MAGIC). The recommendations were produced by a 23 person panel including 21 specialist doctors and researchers and two former COVID-19 patients who were trained to understand the process behind setting up treatment guidelines.
  • Each recommendation was put to a vote. If 80% of the panel agreed, it would be considered a “strong recommendation”. The panel reviewed the combined results from eight randomised control trials — the gold standard for determining cause and effect relationships during which one group receives the drug and another a dummy drug — in which corticosteroids were used as a treatment for COVID-19. These types of studies are called meta studies and they can be conducted when multiple studies address the same question. You can learn more about them from our Epidemiology 101 course. Click here for recordings of the lectures. 
  • Two meta-studies which had already been published on the effects of corticosteroids on COVID-19 were also considered by the panel.

What does the evidence say? 

  • The meta-study included a pooled total of 7 184 patients from the eight randomised control trials.  
  • All eight of the trials considered the effect of receiving corticosteroids on COVID-19 patients in hospital. However, the studies used different types of corticosteroid therapies, including dexamethasone, hydrocortisone and methylprednisolone.
  • The largest study, the RECOVERY trial, randomised 6 425 COVID-19 patients.
    • 2 104 of these patients received dexamethasone while 4 321 received standard care, or the best practice for treating COVID-19 patients.
    • 16% of the RECOVERY trial patients had critical COVID-19 and required the assistance of machines to breathe. 
    • 60% of the study’s patients required supplementary oxygen to be administered. 
    • 24% of the patients didn’t require any additional respiratory intervention.
    • Read more about the RECOVERY trial on our resource page.
  • The seven smaller studies included close to 700 COVID-19 patients considered critically ill and 63 non-critical patients. Half of these patients received some form of corticosteroid intervention while the other half received non-corticosteroid treatments. 
  • One of the eight studies considered (REMAPCAP) had patients from multiple countries including 14 European countries, Australia, Canada, New Zealand, Saudi Arabia and the United Kingdom. The remaining 7 studies were single-country trials:
    • RECOVERY based in the United Kingdom.
    • DEXA-COVID19  and GLUCOCOVID were based in Spain,
    • CoDEX in Brazil,
    • CAPE-COVID in France,
    • COVID STEROID in Denmark, and
    • Steroids-SARI  in China
    • MetCOVID was included as a supplementary study and was based in Brazil. 

Findings from the WHO study:

  • The panel classified COVID-19 patient illness as either critical where patients required ventilation, severe where patients required additional oxygen or non-severe where patients didn’t need either ventilation or oxygen therapies. 
  • In seven of the eight trials fewer critically ill patients had died after 28 days if they were receiving some form of corticosteroid treatment, compared to patients who only received standard care or a placebo.
  • The risk of death of patients with severe COVID-19 was also comparatively reduced. This is based on data from one study.
  • But, COVID-19 patients who didn’t need oxygen or ventilation who received corticosteroid treatments had a bigger risk of dying after 28 days, one of the studies found.

Recommendations from the WHO guidelines:

  • The WHO strongly recommends the use of corticosteroids taken orally or by injection as a treatment for severe and critical COVID-19. This applies to patients whether or not they are hospitalised. The panel is uncertain about the application of the recommendation to children, COVID-19 patients with tuberculosis, and those who are immunocompromised as these groups were underrepresented in the trial studies.
  • The WHO suggests that corticosteroid treatments not be used to treat patients with non-severe COVID-19. This applies whether the patient is hospitalised or not. Patients on chronic corticosteroid treatment should continue their treatment. 
  • Corticosteroids can be administered should the condition of a person who initially does not have severe COVID-19 symptoms worsen. 

The use of corticosteroids for COVID-19 in South Africa:

  • Corticosteroids such as dexamethasone are included in South Africa’s COVID-19 treatment guidelines.The latest version of these guidelines was published on 24 August 2020. South Africa’s guidelines line up with the WHO recommendations. 
  • The department of health suggests that six milligrams of dexamethasone be administered daily for 10 days to COVID-19 patients who are mechanically ventilated or those who need supplemental oxygen but don’t need mechanical breathing assistance. 
  • Should dexamethasone not be available, six milligrams of betamethasone or 40mg of prednisone are proposed as alternatives
  • The country’s recommendations are also based on findings from the RECOVERY trial.
  • The guidelines warn against using dexamethasone in COVID-19 patients who don’t require either additional oxygen or mechanical ventilation. Worth noting, the department highlights, is the potential harm outlined in the RECOVERY trial to patients who didn’t need supplemental oxygen when they were given corticosteroids. These patients experienced a 4% absolute increase in mortality, the department notes (however, these findings were not statistically significant).

Find a review of corticosteroids such as dexamethasone from a subcommittee of South Africa’s Ministerial Advisory Committee on COVID-19 here.
You can find and download the WHO guidelines here.

[Please note: Information on the new coronavirus is rapidly changing. Please refer to the World Health Organisation website for the latest information. Visit for updates on South Africa’s coronavirus response.]

Gopolang Makou was the impact and engagement officer at Bhekisisa.