Objective: The proportion of patients infected with SARS-CoV-2 that are prescribed antibiotics is uncertain, and may contribute to patient harm and global antibiotic resistance. Our objective was to estimate the prevalence and associated factors of antibiotic use in patients with confirmed COVID-19. Methods: We searched MEDLINE, OVID Epub and EMBASE for published literature on human subjects in English up to June 9, 2020. Inclusion criteria were any healthcare settings and age groups; randomized controlled trials; cohort studies; case series with >10 patients; experimental or observational design that evaluated antibiotic prescribing. The main outcome of interest was proportion of COVID-19 patients prescribed an antibiotic, stratified by geographical region, severity of illness, and age. We pooled proportion data using random effects meta-analysis. Results: We screened 7469 studies, from which 154 were included in the final analysis. Antibiotic data were available from 30,623 patients. The prevalence of antibiotic prescribing was 74.6% (95% CI 68.3 to 80.0%). On univariable meta-regression, antibiotic prescribing was lower in children (prescribing prevalence odds ratio (OR) 0.10, 95%CI 0.03 to 0.33) compared to adults. Antibiotic prescribing was higher with increasing patient age (OR 1.45 per 10 year increase, 95%CI 1.18 to 1.77) and higher with increasing proportion of patients requiring mechanical ventilation (OR 1.33 per 10% increase, 95%CI 1.15 to 1.54). Estimated bacterial co-infection was 8.6% (95% CI 4.7-15.2%) from 31 studies. Conclusions: Three-quarters of patients with COVID-19 receive antibiotics, prescribing is significantly higher than the estimated prevalence of bacterial co-infection. Unnecessary antibiotic use is likely high in patients with COVID-19. Registration: PROSPERO (ID CRD42020192286).