Brian M Davis1, Howard Markel2, Alex Navarro2, Eden Wells1, Arnold S Monto1, Allison E Aiello3. 1. Epidemiology Department, University of Michigan School of Public Health. 2. Center for the History of Medicine, University of Michigan Medical School, Ann Arbor. 3. Epidemiology Department, University of North Carolina Gillings School of Global Public Health, Chapel Hill.
Abstract
BACKGROUND: In sum, 559 Michigan schools were closed as a nonpharmaceutical intervention during the influenza A 2009 (H1N1) pandemic. METHODS: By linking the proportion of schools closed within a district to state influenza-like illness (ILI) surveillance data, we measured its effect on community levels of ILI. This analysis was centered by the peak week of ILI for each school district, and a negative binomial model compared three levels of school closure: 0%, 1%-50%, and 51%-100% of schools closed from three weeks leading up to ILI peak to four weeks following ILI peak rate. RESULTS: We observed that school closures were reactive, and there was no statistically significant difference between ILI rates over the study period. There was an elevated rate ratio for ILI at 51%-100% closure, and a reduction in the rate ratio at the 1%-50% compared to the 0% closure level. CONCLUSIONS: These findings suggest that district level reactive school closures were ineffective.
BACKGROUND: In sum, 559 Michigan schools were closed as a nonpharmaceutical intervention during the influenza A 2009 (H1N1) pandemic. METHODS: By linking the proportion of schools closed within a district to state influenza-like illness (ILI) surveillance data, we measured its effect on community levels of ILI. This analysis was centered by the peak week of ILI for each school district, and a negative binomial model compared three levels of school closure: 0%, 1%-50%, and 51%-100% of schools closed from three weeks leading up to ILI peak to four weeks following ILI peak rate. RESULTS: We observed that school closures were reactive, and there was no statistically significant difference between ILI rates over the study period. There was an elevated rate ratio for ILI at 51%-100% closure, and a reduction in the rate ratio at the 1%-50% compared to the 0% closure level. CONCLUSIONS: These findings suggest that district level reactive school closures were ineffective.
Authors: Donne Kofi Ameme; Dora Dadzie; Franklin Asiedu-Bekoe; Elijah Paa Edu-Quansah; Basil Benduri Kaburi; Oxygen Wullar; Patrick Amo-Mensah; Ernest Kenu Journal: BMC Public Health Date: 2020-09-18 Impact factor: 3.295