Measuring the clustering effect of the SARS-CoV-2 transmission in a school population: a cross-sectional study in a high incidence region

Abstract

AbstractSince the beginning of the pandemic, there has been a great deal of controversy regarding the role of schools in the spread of SARS-CoV-2 infection, and the relative contribution of students, teachers, and others. To quantify the clustering effect of SARS-CoV-2 infection within classes and schools considering the seroprevalence of specific antibodies among students and school staff (teachers and non-teachers) evaluated in schools located in the Northern region of Portugal. 1517 individuals (1307 students and 210 school staff) from 4 public and 2 private schools, comprising daycare to secondary levels, were evaluated. A rapid point-of-care test for SARS-CoV-2 specific IgM and IgG antibodies was performed and a questionnaire was completed providing sociodemographic and clinical information. We calculated the seroprevalence of IgM and IgG antibodies and estimated the Median Odds Ratio (OR) and 95% confidence interval (CI) to assess the clustering effect, using a multilevel (school and class) logistic regression. SARS-CoV-2 seroprevalence (IgM or IgG) was 21.8% and 23.8% (p = 0.575) in students and school staff, respectively. A total of 84 (8.6%) students and 35 (16.7%) school staff reported a previous molecular diagnosis. Among students, those who reported high-risk contacts only at school (OR = 1.13; 95% CI 0.72–1.78) had a seroprevalence similar to those without high-risk contacts; however, seroprevalence was significantly higher among those who only reported a high-risk contact outside the school (OR = 6.56; 95% CI 3.68–11.72), or in both places (OR = 7.83; 95% CI 5.14–11.93). Similar associations were found for school staff. The median OR was 1.00 (95% CI 1.00, 1.38) at the school-level and 1.78 (95% CI 1.40, 2.06) at the class-level. SARS-CoV-2 seroprevalence was similar between students and staff, without a clustering effect observed at the school level, and only a moderate clustering effect documented within classes. These results indicate that the mitigation measures in the school environment can prevent the spread of class outbreaks to the remaining school community.