Estimating SARS-CoV-2 Circulation in the School Setting: A Systematic Review and Meta-Analysis

Abstract

Background: Although the contribution of children to SARS-CoV-2 spread is still a matter of debate, school closure has been largely implemented as part of the strategies to mitigate the pandemic. In this meta-analysis, we reviewed the current evidence for SARS-CoV-2 transmission in the educational setting.Methods: We conducted a systematic literature search and review following MOOSE guidelines regarding meta-analysis of observational studies using PUBMED, Ovid MEDLINE, ISI Web of Science and Living Evidence on COVID databases. The search was undertaken during the first week of April, 2021 and repeated on May, 15th 2021 with no restriction by language. Screening, contact tracing and seroprevalence studies were included in the meta-analysis. We analysed data using random-effects meta-regressions.Findings: The searches yielded a total of 1088 publications. In educational setting, we identified 21 screening studies reporting infections, (n>120,000), 15 studies on contact tracing (n=112,622), 9 studies on seroprevalence (n=17,879). From screening studies, we estimated 0·31% (95% CI 0·05-0·81%), point prevalence in schools (I2 = 97%). In contact tracing studies, onward viral transmission was limited (SAR = 2·54%; 95% CI 0·76-5·31%), (I2 = 100%). Summary estimates indicate that young index cases were 74% significantly less likely than adults to favor viral spread (OR = 0·26; 95% CI 0·11-0·63) and data from seroprevalence studies shows that children are 43% less likely than adults to test positive for antibodies (OR = 0·57; 95% CI: 0·49-0·68).Interpretation: This systematic review and meta-analysis suggests that schools did not develop into hotspots for SARS-CoV-2 transmission. With mitigation measures, SARS-CoV-2 viral spread is limited and child-to-adult transmission in the school setting scarce. Young index cases are significantly less likely than adults to favor viral spread and children are significantly less susceptible than adults. The greater percentages of positive people found with contact tracing compared to screening suggests that testing all subjects in schools, independently of symptoms, do not help in reducing clusters.Funding Information: The European Institute of Oncology, Milan, Italy is partially supported by the Italian Ministry of Health with Ricerca Corrente and 5×1,000 funds. Declaration of Interests: We declare no competing interests.