SARS-CoV-2 seroprevalence in children, parents and school personnel from June 2020 to April 2021: cohort study of 55 schools in Switzerland.

Background: Few studies have directly examined the incidence or seroprevalence of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infections in children, parents and teachers from the same school communities. This study aimed to describe SARS-CoV-2 seroprevalence within cantonal districts and school communities in children, parents and school personnel in June-September 2020 and March-April 2021 in the canton of Zürich, Switzerland.

Methods: We invited children from 55 randomly selected primary and secondary schools and 275 classes within them to participate in the Ciao Corona study in June-July 2020. Parents of the participating children and all school personnel were invited in August-September 2020. Eligible classes, parents and school personnel were tested again in March-April 2021. Venous blood was tested for SARS-CoV-2 serology. We collected sociodemographic information of the participants in online questionnaires on enrolment in the study. We excluded vaccinated adults and those with unverified vaccination status from the main analysis. Seroprevalence estimates were adjusted for test accuracy. We assessed the variability of seroprevalence within and across cantonal districts and school communities and compared it with the per capita cumulative incidence of confirmed SARS-CoV-2 infections.

Results: In June-September 2020, 2,473 children, 1,608 school personnel and 2,045 parents participated in the study. In June-September 2020, seroprevalence was low (4.4% to 5.8%) in all cohorts. In March-April 2021, seroprevalence in children and parents (18.1% and 20.9%) was slightly higher than in school personnel (16.9%). We observed a large variation in seroprevalence estimates of the three cohorts within and between districts and school communities, with the median ratio of children's seroprevalence to per capita confirmed cases in district inhabitants of 3.1 (interquartile range 2.6 to 3.9). Seroprevalence was lower in children in the upper school level and their parents, but not teachers. Children's seroprevalence was slightly higher in classes with infected main teachers and families with one infected parent and substantially higher in families with two infected parents.

Conclusions: We observed similar seroprevalence in children and parents, somewhat lower in school personnel in March-April 2021 and striking variation between districts and school communities. Children's seroprevalence was higher in classes with infected main teachers and from families with infected parents.