Estimating the Extent of True Asymptomatic COVID-19 and Its Potential for Community Transmission: Systematic Review and Meta-Analysis

Abstract

Background:The prevalence of true asymptomatic COVID-19 cases is critical to policy makers considering the effectiveness of mitigation measures against the SARS-CoV-2 pandemic. We aimed to synthesize all available research on the asymptomatic rates and transmission rates where possible. Methods:We searched PubMed, Embase, Cochrane COVID-19 trials, and European PMC for preprint platforms such as MedRxiv. We included primary studies reporting on asymptomatic prevalence where: (a) the sample frame includes at-risk population, and (b) there was sufficiently long follow up to identify pre-symptomatic cases. Meta-analysis used fixed effect and random effects models.Results: We screened 571 articles and included five low risk-of-bias studies from three countries (China (2), USA (2), Italy (1)) that tested 9,242 at-risk people, of which 413 were positive and 65 were asymptomatic. Diagnosis in all studies was confirmed using a RT-qPCR test. The proportion of asymptomatic cases ranged from 6% to 41%. Meta-analysis (fixed effect) found that the proportion of asymptomatic cases was 16% (95% CI: 12% -20%) overall; higher in non-aged care 19% (15% -24%), and lower in long-term aged care 8% (4% -14%). Two studies provided direct evidence of forward transmission of the infection by asymptomatic cases but suggested lower rates than symptomatic cases. Conclusion:Our estimates of the prevalence of asymptomatic COVID-19 cases are lower than many highly publicized studies, but still substantial. Further robust epidemiological evidence is urgently needed, including in sub-populations such as children, to better understand the importance of asymptomatic cases for driving spread of the pandemic.

Estimating the burden of COVID-19 on the Australian healthcare workers and health system during the first six months of the pandemic

“…Contributing to this risk, is the fact that asymptomatic persons are potential sources of COVID-19 infection ( Rothe et al, 2020 ). Around 40 to 45% of COVID-19 cases are asymptomatic ( Oran and Topol, 2020 , Byambasuren et al, 2020 ). Whilst we did not have data on whether HCW were furloughed with or without pay, this could also affect presenteeism and outbreak spread.…”

Managing the risk of a COVID-19 outbreak from border arrivals

“…We use a continuous-time branching process to model the spread of COVID-19 (Plank et al, 2020), see Table 1 for parameter values. 33% of infections are assumed to be subclinical and these are 50% as infectious as clinical cases (Byambasuren et al, 2020;Davies, Klepac, et al, 2020;Davies, Kucharski, et al, 2020). Each infected individual causes a randomly generated number ~( , ) of new infections, where = 2.5 for clinical community cases and is the overdisperson parameter (Lloyd-Smith, Schreiber, Kopp, & Getz, 2005).…”

Detection of neutralising antibodies to SARS-CoV-2 to determine population exposure in Scottish blood donors between March and May 2020

“…Symptomatic individuals typically exhibit fever, cough and shortness of breath 2-14 days after infection [5]. However, an unknown proportion of individuals experience no symptoms [6][7][8]. Antibody responses in both symptomatic and asymptomatic individuals are detectable in the blood 14-28 days after infection [9,10].…”