〔Abstract〕 Objective To investigate the clinical improvement of glucocorticoid in the treatment of mild, severe, and critical cases of coronavirus disease 2019(COVID-19). Methods A retrospective study included 248 patients with confirmed COVID-19. The severity of the patient was assessed as mild, severe, and critical. Patients were divided into non-hormonal group and hormonal group according to whether they received glucocorticoid therapy during their hospitalization. The non-hormone group was treated with oxygen support and antiviral therapy, and the hormone group was treated with methylprednisolone for a short term. The discharge mortality rate, hospitalization time, and time from symptom onset to discharge were analyzed. The dynamic changes of S-IgG, S-IgM, S-RBD-IgG, N-IgG, and N-IgM antibodies were detected by ELISA. Results There was no difference in discharge mortality between mild, severe patients in the glucocorticoid therapy group and non-glucocorticoid treatment group, while the mortality was higher in critical patients treated with glucocorticoid(68%vs40%,P<0.05). The hospitalization time in mild patients receiving glucocorticoid therapy was higher than that in the non-glucocorticoid treatment group [22(17-28)dvs16(12-22)d], and the difference was statistically significant(P<0.05). The S-IgG, S-RBD and N-IgG antibodies showed a gradually increasing trend, while the S-IgM and N-IgM antibodies showed a gradually decreasing trend, and there was no statistical significance between the two groups. Conclusion Short-term glucocorticoid therapy can not improve the mortality rate and clinical outcome, and has no significant effect on the immune function of patients with COVID-19.