How school eligibility affects the spread of infectious diseases

A study in JAMA network opened sheds light on how school attendance influences the spread of infectious diseases, using COVID-19 as a case study.

Researchers analyzed the natural age cutoff for kindergarten eligibility in California to compare COVID-19 rates between children who were old enough to attend school and those who were not. This approach, called regression discontinuity, provides a way to quickly understand the role of schools in disease transmission and evaluate the effectiveness of prevention measures within schools without the need for additional data collection or school closures.

The study findings underline the complexity of school-based transmission and highlight the effectiveness of school-based prevention measures. The findings could be applied to other infectious diseases, such as new strains of flu or respiratory viruses, helping schools and public health officials make informed decisions during future outbreaks.

Jennifer Head, senior author of the study, and John G. Searle, assistant professor of epidemiology at the University of Michigan School of Public Health, discuss the study's broader implications for public health and future outbreak responses.

What makes this study's approach to investigating COVID-19 in schools unique?

In this study, we used a method called regression discontinuity to examine the association between school attendance and the number of COVID-19 cases in children. This method looks at whether children born just before the age limit for kindergarten have different COVID-19 rates than children born just afterward.

Children born on either side of this age limit are thought to be similar in most respects, so differences in COVID-19 rates can be attributed, at least in part, to school attendance. While previous studies have used age thresholds for school attendance to study other outcomes such as crime or adult income, to our knowledge this is the first to apply them to understanding the impact of schooling on an infectious disease.

A major advantage of this approach is that it allows us to study the effects of school attendance on the spread of disease without the need for school closures. The approach allows estimation and comparison of effects across different school districts or provinces, allowing identification of district- or provincial-level characteristics (such as vaccination rates) that are associated with weaker associations between schooling and disease transmission.

The approach also uses routine public health data, such as dates of birth and COVID-19 case reports, making it a practical way to analyze school-related disease transmission, even when time and resources are limited.

What does this study reveal?

After taking into account age and the fact that children who attended school were tested more often than those who did not attend school, we found that COVID-19 rates were higher among children eligible for kindergarten compared to those who did not during the in-person semesters.

Importantly, this difference narrowed with each semester, and there was lower transmission among children who were eligible for school compared to children who were not eligible during the summer break after in-person semesters.

Specifically, COVID-19 rates were 52% higher for school-eligible children than for ineligible children in the fall semester of 2021, 26% higher in the spring semester of 2022, and 19% higher in the fall of 2022. When schools were closed for in-person education, we found no difference in COVID-19 rates between the two groups, supporting the idea that these differences are more likely to be related to school attendance than with other factors.

We also found no association between in-person instruction and COVID-19 hospitalizations. This is consistent with other studies showing that school attendance has little impact on COVID-19 hospitalizations, likely because severe cases are relatively rare in young children.

How can the findings of this study guide schools' approach to future outbreaks of infectious diseases?

Wherever people gather, the risk of spreading infectious diseases is greater, and schools are no exception. In our study, we saw higher transmission of COVID-19 among school-aged children compared to children who were not eligible for school.

That said, it is critical to weigh this risk against the significant downsides of closing schools, such as learning loss, mental health issues and wider disparities in education and the workplace.

In this study, when we compared the association between school eligibility and COVID-19 with other settings, such as social gatherings outside of school, we found that the risk was not greater in schools. Based on the strength of our collaboration, we believe that our findings support the use of in-school warning measures as much as possible during school closures.

Although children are less likely than adults to experience severe symptoms of COVID-19, other illnesses, such as influenza, can have more serious consequences for children.

The method we used in this study provides a rapid and effective way to evaluate how in-person schooling affects the spread of different pathogens, especially for emerging pathogens or new strains of known pathogens. This method may also be useful to investigate the extent to which childhood vaccinations reduce associations with school illnesses, which is especially important in the context of increasing vaccine hesitancy.

What role did vaccination and natural immunity play in the changes you observed over time?

COVID-19 vaccinations for children ages 5 to 11 became available late in the fall 2021 semester, meaning children on both sides of the kindergarten age cutoff could receive vaccine-induced protection by the start of the spring 2022 semester.

Over time, we saw a steady decline in the association between school eligibility and COVID-19 rates, likely reflecting growing immunity to both vaccination and natural infections in children.

One piece of evidence supporting this theory is our finding that COVID-19 rates among school-aged children were lower than among school-aged children during the 2022 summer holidays. One explanation is that school-age children developed higher levels of natural immunity during the summer holidays. the 2021-2022 school year, which helped protect them over the summer.

Second, we found that counties with higher vaccination rates among children and adults had weaker associations between school eligibility and COVID-19 rates. Although this trend was not statistically significant, it was consistent across all three in-person semesters.

Third, when comparing counties, we found that larger counties — home to California's largest school districts, such as Los Angeles and San Francisco — had a weaker association between education eligibility and COVID-19 rates.

These counties had higher transmission before in-person schooling began, which may have increased natural immunity among children. They also had higher vaccination rates and followed stricter mask mandates, which likely contributed to this trend.

What else does this research tell us about the spread of disease in schools?

The link between school eligibility and COVID-19 rates likely reflects more than just what's happening in the classroom. Other factors related to schooling – such as riding the bus, participating in sports or after-school programs, or getting a parent back to work – can also play a role.

This is important because previous research suggests that children's social interactions outside the classroom can lead to higher transmission rates than interactions inside the classroom.

Presented by the University of Michigan School of Public Health