The rapid spread of severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) caused the coronavirus disease 2019 (COVID-19) pandemic. Several pharmaceutical and non-pharmaceutical measures have been implemented to contain the pandemic. Among the non-pharmaceutical approaches, school closure has been widely implemented to reduce further transmission of SARS-CoV-2.
Study: No causal effect of school closures in Japan on the spread of COVID-19 in spring 2020. Image Credit: Kathryn Sullivan/ Shutterstock
Background
Most schools closed soon after the World Health Organization announced COVID-19 infection to be a pandemic. This strategy affected 84.3% of the world’s enrolled students. Some of the impacts of school closure strategy include learning loss, deterioration of physical health, etc. Considering all the negative impacts, it is imperative to know if the benefits of school closures outweigh the adverse effects. To date, scientists have been unable to draw a robust conclusion of the degree of the benefit (if any) of closing schools.
Previous empirical studies have reported that school closures had effectively reduced the transmission of SARS-CoV-2. However, others failed to show any statistically significant effects of this policy. Scientists believe one of the reasons for contradictory findings is the differential methodological approach. For example, one of the drawbacks of simulation studies is that it assumes parameters for their models and, thereby, the values may not be real. Similarly, most empirical studies estimate parameters using publicly available aggregated data, and these data might not be rigorous to determine causal inference.
A typical research design is panel regression utilizing a large dataset across countries and days. In previous studies, researchers regressed the number of cases on a dummy variable to predict the effect of school closures. The effect of school closure was estimated by measuring the change in the number of COVID-19 cases between days when a country closed and opened its schools. However, the majority of studies did not control for any other variables. Thereby, dozens of potential confounders were not accounted for, which could yield biased results. Also, the possibility of reverse causality (the decision of closing schools was due to a high number of COVID-19 cases) exists.
A new study
A new study published in Nature Medicine estimated the causal effect of school closures on reducing the spread of COVID-19. In this study, researchers used data from Japan, where a few municipalities implemented school closure while others did not. This variation in school closure was studied among hundreds of municipalities by utilizing matching techniques considering dozens of confounders.
The results obtained in this study suggested that the effect of school closures on COVID-19 cases in Japan in early 2020 was not significantly different from zero. The authors stated that although the research design could not elucidate the mechanisms underlying the empirical findings, it provided a few conjectures. Firstly, some treated municipalities did not comply with government rule completely, i.e., although the schools remained closed, they provided spaces such as playing fields, gyms, classrooms, and libraries for children of working parents. Also, children could have interacted with each other outside school.
Secondly, similar to previous studies, in control municipalities, children may be less susceptible to SARS-CoV-2 infection and might not transmit the virus to others, including parents, teachers, and neighbors. Additionally, students who were attending in-person schooling followed strict mitigation measures such as physical distancing and enhanced hygiene. This would limit interactions between students. Thereby, closing down schools might not have had a significant impact on reducing the transmission of SARS-CoV-2 infection.
Limitations
This study had certain limitations. For instance, the authors stated that even though they could validate the internal findings, they were less confident about the external validity. Furthermore, unlike many countries such as Italy, Brazil, Spain, and the United States, Japan was less impacted by COVID-19. Therefore there was a possibility that school closures only had an apparent effect on COVID-19 cases once the reproduction number within a school setting passed a definite threshold.
Although researchers determined the effectiveness of school closures in the early months of the pandemic, community transmission has occurred rapidly since spring 2020. The emergence of SARS-CoV-2 variants (Alpha, Delta, etc.) with increased transmissibility could also explain the findings. The behavior of the citizens in obeying the rules, such as social distancing, using facemasks, etc., could have also affected the results, but these factors were not appropriately considered.
Conclusion
The current study offered both empirical and methodological contributions to the existing literature. The authors found no empirical evidence that school closures in Japan significantly reduced the number of coronavirus cases. Therefore, in the future, policymakers must be cautious when considering similar policies as it affects the well-being of both children and parents. Researchers recommended that governments should assess the SARS-CoV-2 infection rates and school closures at a granular level, i.e., municipality or school district, daily. They developed a method after considering the causal inference by using matching techniques. This methodology also exploited features of Japanese municipalities that helped identify the effect of school closures, independent of other non-pharmaceutical interventions.
One in four healthy young Marines report long-term physical, cognitive, or psychiatric effects after mild COVID-19