Objective To systematically review the efficacy and safety of glucocorticoids for severe COVID-19 and to provide references for the treatment strategy of severe COVID-19 patients. Methods PubMed, EMbase, The Cochrane Library, Web of Science, ClinicalTrials.gov, WanFang Data and CNKI databases were electronically searched to collect randomized controlled trials (RCTs) that reported glucocorticoid therapy for severe COVID-19 patients from inception to August 26th, 2021. Two reviewers independently screened the literature, extracted data, and assessed the risk of bias of the included studies. Meta-analysis was then performed using RevMan 5.3 software. Results A total of 7 RCTs involving 6 236 patients were included. The meta-analysis results showed that compared with usual care, glucocorticoids significantly reduced the all-cause mortality of severe COVID-19 (RR=0.84, 95%CI 0.77 to 0.91, P<0.000 1), whereas no significant difference was found in the progression of complex diseases between the two groups (RR=0.84, 95%CI 0.69 to 1.01, P=0.06). Glucocorticoids did not increase adverse effects in severe COVID-19 compared with usual care (general adverse events: RR=1.15, 95%CI 0.66 to 2.03, P=0.62; serious adverse events: RR=1.13, 95%CI 0.54 to 2.38, P=0.75). Conclusion Current evidence suggests that glucocorticoids are effective in treating severe COVID-19 without significantly increasing adverse events. However, due to the limited quantity and quality of the included studies, more high-quality studies are needed to verify the conclusion.