Steroids are expected to be effective in the treatment of cytokine release syndrome, which is considered to be associated with severe cases of coronavirus disease 2019 (COVID-19). We aimed to investigate the use of steroids and its effects. We conducted a retrospective chart review and an analysis of 226 consecutive hospitalized patients with confirmed COVID-19. Patients were divided into those who received steroids (steroid group) and those who did not (no steroid group). Inverse probability weighted analysis was performed to assess the effect of steroids on in-hospital mortality. The steroid group had higher rates of preexisting hypertension and peripheral vascular disease as well as higher lactate dehydrogenase levels, d-dimer levels, and inflammatory markers than the no steroid group (all P <0.05). The steroid group had significantly higher rates of multifocal pneumonia than the no steroid group at admission (75.4% vs. 50.3%, P = 0.001). Notably, the steroid group had higher rates of developing bacterial infection (25% vs. 13.1%, P = 0.041) and fungal infection (12.7% versus 0.7%, P <0.001) during the hospital course than the no steroid group. After adjustment, it was observed that steroids did not decrease or increase in-hospital mortality (odds ratio [95% confidence interval]: 1.02 [0.60–1.73, P = 0.94]). There was an increase in bacterial and fungal infections with steroid use.