To prevent the spread of COVID-19 at school, start with local data and calculate

As schools across the nation continue to work on constructing online and in-class learning, a new study at the UCSF Benioff Children’s Hospital requires that published numbers and simple data are needed to estimate the number of potentially infected students. It’s just an equation. In the classroom. The results may help the district make hard decisions about whether and how many students are safe to return to the classroom.

In a study published in JAMA Pediatrics A comparison of the prevalence of SARS-CoV-2, the virus that causes COVID-19, among asymptomatic children on August 25, 2020, was confirmed in a region of asymptomatic children. It turned out to roughly reflect the number of cases. General population..

Researchers tracked SARS-CoV-2 test results for more than 33,000 children at 28 hospitals in 25 regions across the country. They were regularly tested before undergoing surgery or non-viral medical treatment. About 250 patients were found to have asymptomatic COVID. These patients accounted for 0.65% of the total Pediatric patient 0% to 2.2% tested, or 1/155 depending on region.

Consistent linear relationships across all regions

The researchers then compared each hospital’s data with weekly confirmed cases in the hospital catchment available at the Johns Hopkins Coronavirus Research Center. “We found a linear relationship between the prevalence of SARS-CoV-2 infection in the asymptomatic pediatric population and cases in the general population, and this association was found to be consistent across all regions.” Said senior author Dylan K. Chan. MD, Ph.D. From the UCSF Department of Pediatric Otolaryngology. “In regions where the general population had twice the prevalence of COVID compared to other regions, there was twice as much the rate of asymptomatic childhood infections.”

The researchers reversed this procedure and calculated the prevalence of asymptomatic children using published data and a developed formula (prevalence of asymptomatic children (%) = 1.07 x week). Incidence rate (per 1,000 general population) + 0.23). “Substituting another set of published data into the equations to generate “estimated” prevalence,” Chan said. “In addition, the estimates proved to be fairly accurate when compared to the latest prevalence rates obtained for an additional 15,000 children who were asymptomatic.”

As an example, the prevalence of subclinical COVID in patients tested at the UCSF Benioff Children’s Hospital (one of 28 hospitals in the study) was 0.64% in June. The watershed of the hospital covers the cities of San Francisco, Auckland, San Jose and the surrounding 14 counties. The researchers used cases identified weekly in the same period with public data from this catchment and applied those equations to reach 0.79%. This was within the statistical error of the actual measurement.

Chan uses this formula to Classroom.. “A wide area test is impractical, as we need to test thousands of children in the area to get an estimate of asymptomatic infections,” said Chan in a single population of 1,000. Note that the data from makes it impossible to make an accurate estimation.

San Francisco school risk reaches 30%

Based on published data in the first half of August, researchers estimate that asymptomatic childhood prevalence in San Francisco is 1.1%. In a hybrid model classroom with 11 students, at least one asymptomatic student has an 11% chance of being infected, even if all students were screened for COVID-19 symptoms and asymptomatic. is. In a normal lower grade classroom with 22 students, the risk increases to 22%, and in a normal higher grade class with 33 students, the risk reaches 30%.

In Merced County, California’s Central Valley, the estimated pediatric morbidity is 4.6%, with at least one infected student in a hybrid model classroom with at least 11 students and 64% in a 22 student classroom. , 79 Percentage of 33 students in the classroom. In contrast, the estimated childhood prevalence in New York is 0.7% and at least one person may be infected student In the classroom, 8%, 15%, and 21% got nailed, respectively.

While the risk of an asymptomatic child infecting another child remains unknown, the model uses continuously updated public data to identify children with active but asymptomatic COVID. It may represent the best way to estimate the percentage. As the team developed the model using 25 different regions across countries, researchers believe their findings should be generalizable to most communities across the country.