Abstract
This meta-analysis aims to synthesize global evidence on the risk of reinfection among people previously infected with SARS-CoV-2. We systematically searched PubMed, Scopus, Embase and Web of Science as of April 5, 2021. We conducted: (1) meta-analysis of cohort studies containing data sufficient for calculating the incidence rate of SARS-CoV-2 reinfection; (2) systematic review of case reports with confirmed SARS-CoV-2 reinfection cases. The reinfection incidence was pooled by zero-inflated beta distribution. The hazard ratio (HR) between reinfection incidence among previously infected individuals and new infection incidence among infection-naïve individuals was calculated using random-effects models. Of 906 records retrieved and reviewed, 11 studies and 11 case reports were included in the meta-analysis and the systematic review, respectively. The pooled SARS-CoV-2 reinfection incidence rate was 0.70 (standard deviation [SD] 0.33) per 10,000 person-days. The incidence of reinfection was lower than the incidence of new infection (HR = 0.12, 95% confidence interval 0.09-0.17). Our meta-analysis of studies conducted prior to the emergency of the more transmissible Omicron variant showed that people with a prior SARS-CoV-2 infection could be re-infected, and they have a lower risk of infection than those without prior infection. Continuing reviews are needed as the reinfection risk may change due to the rapid evolution of SARS-CoV-2 variants.