Prednisone can be an effective treatment for COVID-19, but only in a specific subset of patients: those sick enough to need supplemental oxygen or mechanical ventilation. For people with mild symptoms who are managing the illness at home, prednisone and other corticosteroids offer no benefit and may actually increase the risk of death. This distinction between severe and mild disease is not a minor caveat. It is the single most important thing to understand about steroids and COVID-19, and getting it wrong has real consequences.
What the Landmark Trials Actually Showed
The strongest evidence for corticosteroids in COVID-19 comes from the RECOVERY trial, a massive study conducted in the United Kingdom. Among hospitalized patients randomized to receive dexamethasone (a corticosteroid closely related to prednisone), 28-day mortality dropped from about 26% to 23% overall. But the benefit was not evenly distributed. Patients on mechanical ventilation saw the largest reduction: death rates fell from roughly 41% to 29%. Patients receiving supplemental oxygen without ventilation also benefited, with mortality dropping from about 26% to 23%. However, among patients who were hospitalized but did not need any respiratory support, dexamethasone appeared to increase the risk of dying, with mortality trending from 14% in the usual-care group to nearly 18% in the steroid group.1PubMed. Dexamethasone in Hospitalized Patients with Covid-19
A WHO-sponsored meta-analysis pooling data from seven randomized trials of critically ill COVID-19 patients confirmed the direction of these findings. Corticosteroids reduced mortality with a summary odds ratio of about 0.66, meaning roughly a third fewer deaths compared to usual care or placebo.2JAMA. Association Between Administration of Systemic Corticosteroids and Mortality Among Critically Ill Patients With COVID-19: A Meta-analysis A broader systematic review covering 22 studies of hospitalized patients with moderate or severe respiratory failure found a similar benefit, with an overall odds ratio of 0.72.3PubMed Central. Corticosteroid use in COVID-19 patients: a systematic review and meta-analysis on clinical outcomes
Why Prednisone Instead of Dexamethasone
Most of the headline-making trials used dexamethasone, not prednisone. This raises a reasonable question: does the evidence even apply to prednisone? In practice, yes. Prednisone, dexamethasone, methylprednisolone, and hydrocortisone are all systemic corticosteroids that work through the same core mechanism, dampening the inflammatory cascade that drives severe COVID-19 lung injury.4Journal of Intensive Medicine. Corticosteroids for COVID-19 Treatment guidelines from the National Institutes of Health list dexamethasone as the preferred agent, but explicitly name prednisone as an acceptable alternative at equivalent doses when dexamethasone is unavailable.5PubMed Central. Corticosteroids for treatment of COVID-19: effect, evidence, expectation and extent
In some hospital protocols, patients initially treated with intravenous methylprednisolone were later transitioned to oral prednisone for the remainder of their course.6PLOS ONE. Dexamethasone vs methylprednisolone high dose for Covid-19 pneumonia Dexamethasone tends to be the default in hospital settings because of its long duration of action and the simplicity of once-daily dosing, but prednisone is more commonly prescribed in outpatient settings since it comes in oral tablets at various strengths. The key point is that corticosteroids as a class have proven benefit in severe COVID-19. The specific molecule matters less than whether the patient actually needs one.
The Danger of Using Steroids Too Early or in Mild Cases
COVID-19 unfolds in two overlapping phases. During the first week or so, the virus is actively replicating. After that, in some patients, the immune system overreacts, and it is this exaggerated inflammatory response that destroys lung tissue and can become fatal. Corticosteroids suppress the immune system. That is precisely why they help during the inflammatory phase, and precisely why they can backfire during the viral phase, when you need your immune system working at full strength to clear the virus.
A meta-analysis focused specifically on hospitalized COVID-19 patients who were not receiving supplemental oxygen found that corticosteroid use was associated with significantly higher mortality. Among steroid-treated patients, about 14% died, compared to 10% in the control group, with a number needed to harm of 27. In other words, for roughly every 27 patients in this category who received steroids unnecessarily, one additional patient died.7PubMed. Meta-Analysis of Glucocorticoids for Covid-19 Patients Not Receiving Oxygen A systematic review of noncritically ill patients similarly found that steroid-treated patients experienced longer hospital stays and prolonged viral shedding compared to those who received no steroids.8Scientific Reports. A systematic review of corticosteroid treatment for noncritically ill patients with COVID-19
For outpatients with mild symptoms, the evidence is even less encouraging. A randomized trial of steroids in outpatient COVID-19 found no significant difference in hospitalization rates; if anything, the steroid group had a slightly higher rate of admission, though the numbers were small.9PubMed Central. Steroid in the Treatment of Outpatient COVID-19: A Multicenter Randomized Controlled Trial Taking prednisone at home because you feel lousy with COVID is not supported by the data and can make things worse.
Timing Within the Hospital Stay
Even among patients sick enough to warrant steroids, when those steroids start matters. A multicenter retrospective study found that patients who received their first steroid dose more than seven days after symptom onset had significantly lower mortality, with a hazard ratio of 0.56 compared to those who received steroids earlier. The same study found a similar benefit for patients whose steroids began more than 72 hours into their hospitalization.10PubMed Central. Timing of corticosteroids impacts mortality in hospitalized COVID-19 patients Another multicenter study confirmed this pattern, showing that steroid initiation eight or more days after symptom onset was associated with lower mortality.11PubMed Central. Steroid initiation timing and outcome of coronavirus disease 2019 patients
The logic tracks with the two-phase model of the disease. Giving steroids during active viral replication suppresses the very immune response that is fighting the virus. A case report published in BMJ Case Reports illustrates this risk: immunosuppression during the early viral phase can predispose patients to more severe disease rather than protecting them from it.12PubMed. Steroid harms if given early in COVID-19 viraemia The practical implication is that clinicians should not reflexively reach for steroids the moment a COVID-19 patient is admitted. The trigger should be worsening respiratory status combined with appropriate timing relative to symptom onset, not simply a positive test result.
How Steroids Affect Viral Clearance
One concern with immunosuppressive therapy during an active infection is that it might slow down the body’s ability to eliminate the virus. The evidence here is mixed but tilts toward caution. A systematic review and meta-analysis found that corticosteroid use was associated with delayed viral clearance, with patients roughly 50% more likely to continue shedding virus compared to those not treated with steroids. The effect was dose-dependent: high and medium doses carried greater risk of delayed clearance, while low doses did not show a statistically significant delay.13PubMed Central. Association between glucocorticoids treatment and viral clearance delay in patients with COVID-19: a systematic review and meta-analysis
On the other hand, a single-center study found that early corticosteroid treatment did not significantly affect viral clearance in patients with moderate or severe disease, with factors like age, symptom duration at admission, and baseline oxygen levels mattering more.14Scientific Reports. Viral clearance after early corticosteroid treatment in patients with moderate or severe covid-19 The takeaway is that delayed clearance is a real phenomenon, particularly at higher doses, but in patients who genuinely need steroids to survive the inflammatory storm, a few extra days of detectable virus is an acceptable trade-off.
Side Effects That Clinicians Watch For
Corticosteroids are not benign drugs, and the side effects in COVID-19 patients have been well documented. Some of the most consequential include:
- Blood sugar spikes: Corticosteroids reliably raise blood glucose levels, and this effect is more severe in patients who already have diabetes. A study at an academic medical center found that patients with diabetes required significantly higher doses of corrective insulin while on steroid therapy.15PubMed Central. Burden of Hyperglycemia in Patients Receiving Corticosteroids for Severe COVID-19 In some countries, steroid-induced hyperglycemia in COVID-19 patients contributed to rising cases of diabetic ketoacidosis and mucormycosis, a dangerous fungal infection.16PubMed Central. Diagnosis and Management Considerations in Steroid-Related Hyperglycemia in COVID-19
- Secondary fungal infections: The immunosuppressive effect of steroids that makes them useful against the cytokine storm also leaves patients vulnerable to opportunistic infections. Longer steroid courses have been associated with an increased risk of fungal co-infections.17PubMed Central. Fungal infection-related conditions and outcomes in severe COVID-19 This was described as a “double-edged sword” in a review of secondary fungal infections in COVID-19 patients.18PubMed Central. When to Initiate Antifungal Treatment in COVID-19 Patients with Secondary Fungal Co-infection
- Psychiatric effects: Steroids can cause mood changes, insomnia, agitation, and in rare cases, acute psychosis. A case study documented a patient who developed persecutory delusions and hallucinations after receiving intravenous methylprednisolone, symptoms that resolved after the steroid dose was reduced and an antipsychotic was started.19European Psychiatry. Acute psychosis following corticosteroid administration for COVID-19 and Respiratory Syncytial Virus infection
These risks reinforce why steroids should be reserved for patients whose clinical situation demands them. The mortality benefit in severe disease is substantial enough to justify these side effects, but the calculus changes entirely when the patient is not critically ill.
Rebound Inflammation After Stopping Steroids
Some patients experience a resurgence of symptoms after their steroid course ends, a phenomenon sometimes called rebound pneumonia. A study analyzing this pattern found that patients in the rebound group tended to have more severe respiratory failure at baseline and showed higher inflammatory markers at the time steroids were stopped. The researchers identified a prediction model using age over 68, high supplemental oxygen requirements, low lymphocyte counts, and elevated immune markers to identify patients at higher risk of rebound.20BMJ Open Respiratory Research. Clinical features of COVID-19 patients with rebound phenomenon after corticosteroid therapy
When rebound does occur, the standard approach has been to restart corticosteroids at a higher dose and for a longer duration than the original course.21PubMed Central. A Case of Rebound Inflammation in a 38-Year-Old Man with Severe COVID-19 Pneumonia Following Cessation of Dexamethasone Therapy This is one reason clinicians sometimes taper steroids gradually rather than stopping them abruptly, especially in patients who were severely ill. If you have been prescribed a steroid course for COVID-19 and are told to taper rather than simply stop, this is why.
Combining Steroids With Other Treatments
Corticosteroids became the standard of care for hospitalized patients with severe or critical COVID-19, and they are now the backbone on which other immunomodulatory treatments are layered.22PubMed Central. COVID-19: A Review of Potential Treatments Tocilizumab, an interleukin-6 receptor blocker, and baricitinib, a JAK inhibitor, are both recommended for use in combination with a corticosteroid in certain hospitalized patients. A systematic review and meta-analysis found that adding tocilizumab to corticosteroids reduced mortality odds by about 22% compared to corticosteroids alone. Baricitinib showed a similar reduction, though the evidence was somewhat less certain.
The idea behind these combinations is that corticosteroids provide broad-spectrum suppression of the inflammatory response, while drugs like tocilizumab and baricitinib target specific immune pathways that drive the most damaging aspects of the cytokine storm. For patients on ventilators or progressing despite steroids alone, these add-on therapies have become a standard consideration.
Inhaled Steroids as an Alternative
A separate and somewhat more encouraging line of research has explored inhaled corticosteroids, the kind of steroid found in asthma inhalers. Because inhaled steroids act locally in the airways rather than flooding the whole body, they carry fewer systemic side effects. A systematic review and meta-analysis of randomized trials found that patients receiving inhaled corticosteroids had significantly shorter hospital stays, about a day and a half less on average, compared to those receiving placebo.23PubMed Central. Inhaled corticosteroids’ effect on COVID-19 patients: A systematic review and meta-analysis of randomized controlled trials
Inhaled steroids occupy a different niche than systemic drugs like prednisone. They are being studied mainly for early or mild-to-moderate disease, where systemic steroids have shown either no benefit or harm. The appeal is that they could reduce inflammation in the lungs specifically, without broadly suppressing the immune system’s ability to fight the virus elsewhere. This area is still developing, but inhaled budesonide in particular has attracted attention as a potential outpatient treatment for early COVID-19.
Steroids and Long COVID
Whether steroids can help with persistent symptoms after the acute infection, often called long COVID, is a question with much less settled evidence. A small study of eight patients with post-COVID syndrome who received a short four-day course of oral prednisone found that all described complete or partial recovery of their symptoms. Joint pain improved in about 86% of treated patients, muscle pain in 75%, and fatigue in more than half.24PubMed Central. A Short Corticosteroid Course Reduces Symptoms and Immunological Alterations Underlying Long-COVID Another study of patients with persistent lung abnormalities and low oxygen levels found that a longer course of steroids tapered over 8 to 10 weeks improved breathlessness in the majority and normalized CT scans in a quarter.25Monaldi Archives for Chest Disease. Systemic corticosteroids for management of ‘long-COVID’: an evaluation after 3 months of treatment
These are small, preliminary studies, and they are far from the kind of large randomized evidence that supports steroid use in acute severe COVID-19. A proper randomized controlled trial, PreVitaCOV, has been designed to test prednisolone in primary care patients with post-COVID syndrome, but results are still pending.26PubMed Central. Feasibility, safety and effectiveness of prednisolone and vitamin B1, B6, and B12 in patients with post-COVID-19-syndrome (PreVitaCOV) Until larger trials report, treating long COVID with prednisone remains largely a clinical judgment call rather than a guideline-backed recommendation. The hypothesis that ongoing low-grade inflammation drives some long COVID symptoms is plausible and supported by early biological data, but the evidence is still thin enough that clinicians are rightly cautious.
Lessons From Earlier Coronavirus Outbreaks
Before COVID-19, the role of steroids in respiratory coronavirus infections was genuinely controversial. During the SARS and MERS outbreaks, steroids were used widely but without strong trial evidence. A systematic review spanning all three coronavirus diseases found that when all studies were pooled without adjusting for disease severity, steroids showed no overall mortality benefit. But once the analysis was restricted to studies that properly accounted for how sick patients were at baseline, the adjusted results showed a significant benefit, with a hazard ratio of about 0.38.27PubMed Central. Efficacy of Corticosteroids in Patients with SARS, MERS and COVID-19: A Systematic Review and Meta-Analysis
This earlier confusion partly explains why health authorities were initially hesitant to recommend steroids for COVID-19. The RECOVERY trial in mid-2020 changed the conversation decisively, but the earlier ambiguity highlights an important point: steroids look harmful when given indiscriminately to all patients, including those with mild disease, and they look beneficial when given selectively to those with serious inflammatory lung injury. The lesson from SARS and MERS turned out to apply directly to COVID-19, but it took a well-designed large trial to make it clear.
Children and Multisystem Inflammatory Syndrome
In children, acute COVID-19 is typically mild and does not require steroid treatment. The situation changes with multisystem inflammatory syndrome in children (MIS-C), a rare but serious post-infectious condition in which the immune system attacks multiple organ systems, sometimes including the heart. Treatment for MIS-C involves intravenous immunoglobulin as a first-line therapy, with corticosteroids considered as part of the initial management. In refractory cases, more targeted immune therapies may be added. The corticosteroid role in MIS-C is distinct from its role in adult COVID-19 pneumonia: it is addressing a different immune process entirely, one that resembles Kawasaki disease more than viral pneumonia.