ABSTRACT
The association between current smoking and coronavirus disease 2019 (COVID-19) progression remains uncertain. We aim to provide up-to-date evidence of the role of cigarette smoking in COVID-19 hospitalisation, severity and mortality. On 23 February 2022 we conducted an umbrella review and a traditional systematic review via PubMed/Medline and Web of Science. We used random-effects meta-analyses to derive pooled odds ratios of COVID-19 outcomes for smokers in cohorts of severe acute respiratory syndrome coronavirus 2 infected individuals or COVID-19 patients. We followed the Meta-analysis of Observational Studies in Epidemiology reporting guidelines. PROSPERO: CRD42020207003. 320 publications were included. The pooled odds ratio for current versus never or nonsmokers was 1.08 (95% CI 0.98-1.19; 37 studies) for hospitalisation, 1.34 (95% CI 1.22-1.48; 124 studies) for severity and 1.32 (95% CI 1.20-1.45; 119 studies) for mortality. Estimates for former versus never-smokers were 1.16 (95% CI 1.03-1.31; 22 studies), 1.41 (95% CI: 1.25-1.59; 44 studies) and 1.46 (95% CI 1.31-1.62; 44 studies), respectively. Estimates for ever- versus never-smokers were 1.16 (95% CI 1.05-1.27; 33 studies), 1.44 (95% CI 1.31-1.58; 110 studies) and 1.39 (95% CI 1.29-1.50; 109 studies), respectively. We found a 30-50% excess risk of COVID-19 progression for current and former smokers compared with never-smokers. Preventing serious COVID-19 outcomes, including death, seems the newest compelling argument against smoking.