Reliability of Early Estimates of the Basic Reproduction Number of COVID-19: A Systematic Review and Meta-Analysis

Objective: This systematic review estimated the pooled [Formula: see text] for early COVID-19 outbreaks and identified the impact of study-related factors such as methods, study location and study period on the estimated [Formula: see text]. Methods: We searched electronic databases for human studies published in English between 1 December 2019 and 30 September 2020 with no restriction on country/region. Two investigators independently performed the data extraction of the studies selected for inclusion during full-text screening. The primary outcome, [Formula: see text] , was analysed by random-effects meta-analysis using the restricted maximum likelihood method. Results: We identified 26,425 studies through our search and included 151 articles in the systematic review, among which 81 were included in the meta-analysis. The estimates of [Formula: see text] from studies included in the meta-analysis ranged from 0.4 to 12.58. The pooled [Formula: see text] for COVID-19 was estimated to be 2.66 (95% CI, 2.41–2.94). The results showed heterogeneity among studies and strong evidence of a small-study effect. Conclusions: The high heterogeneity in studies makes the use of the [Formula: see text] for basic epidemic planning difficult and presents a huge problem for risk assessment and data synthesis. Consensus on the use of [Formula: see text] for outbreak assessment is needed, and its use for assessing epidemic risk is not recommended.