Effects of school-level and area-level socio-economic factors on elementary school student COVID-19 infections: a population-based observational study

OBJECTIVES: To estimate the variability of the cumulative incidence of SARS-CoV-2 infections among elementary school students attributable to individual schools and/or their geographic areas, and to ascertain whether socio-economic characteristics of school populations and/or geographic areas may be predictive of this variability. DESIGN: Population-based observational study of SARS-CoV-2 infections among elementary school children. SETTING: 3994 publicly funded elementary schools in 491 forward sortation areas (designated geographic unit based on first three characters of Canadian postal code), Ontario, Canada, September 2020 to April 2021. PARTICIPANTS: All students attending publicly funded elementary schools with a positive molecular test for SARS-CoV-2 reported by the Ontario Ministry of Education. MAIN OUTCOME MEASURES: Cumulative incidence of laboratory-confirmed elementary school student SARS-CoV-2 infections in Ontario, 2020–21 school year. RESULTS: A multilevel modelling approach was used to estimate the effects of socio-economic factors at the school and area levels on the cumulative incidence of elementary school student SARS-CoV-2 infections. At the school level (level 1), the proportion of the student body from low-income households was positively associated with cumulative incidence (β=0.083, p<0.001). At the area level (level 2), all dimensions of marginalisation were significantly related to cumulative incidence. Ethnic concentration (β=0.454, p<0.001), residential instability (β=0.356, p<0.001) and material deprivation (β=0.212, p<0.001) were positively related, while dependency (β=-0.204, p<0.001) was negatively related. Area-related marginalisation variables explained 57.6% of area variability in cumulative incidence. School-related variables explained 1.2% of school variability in cumulative incidence. CONCLUSIONS: The socio-economic characteristics of the geographic area of schools were more important in accounting for the cumulative incidence of SARS-CoV-2 elementary school student infections than individual school characteristics. Schools in marginalised areas should be prioritised for infection prevention measures and education continuity and recovery plans.