The role of children in transmission of SARS-CoV-2

by Gertrud U. Rey

It is well established that children experience less severe disease after infection with SARS-CoV-2. However, to what extent infected children contribute to transmission of the virus is less clear. This topic is of great interest as we prepare for the start of a new school year. 

Are children less susceptible to infection with SARS-CoV-2 or are they just less likely to develop clinical symptoms? Do infected children produce as much virus as infected adults? How likely are asymptomatically infected children to transmit the virus to others? The answers to these questions are inconsistent and confusing, for at least a couple of reasons. At this time it is difficult to directly measure these variables because most schools have been closed for the better part of the pandemic and children have been isolated from others. Also, children are often missed as index (first identified) cases in groups of related cases because they are more likely to be asymptomatic. To assess transmission from children to adults, many scientists have turned to statistical models. 

The authors of one such published statistical study applied data from China, Japan, Italy, Singapore, South Korea, and Canada to three different variants of a mathematical model to analyze age-dependent effects on transmission of SARS-CoV-2. A first variant of the model assumed that susceptibility to infection varies by age – with susceptibility being the probability of infection on contact with an infected person; a second variant assumed that the likelihood of developing clinical symptoms varies by age; and a third variant assumed that neither susceptibility nor clinical disease depend on age. The authors conclude that people under the age of 20 are half as susceptible to infection as those over 20, and that interventions aimed at children have at most a 20% impact on reducing SARS-CoV-2 transmission.

Another modeling study (a preprint, not yet peer-reviewed) out of Israel suggests that children are slightly less than half as susceptible to infection as adults, while their infectivity is 85% relative to that of adults. Infectivity is the probability that an individual can become infected within a certain time frame when making contact with another, identically susceptible individual. The authors conclude that children are less likely to become infected with SARS-CoV-2 than adults and that the chances of infection increase with age. 

A preprint out of France suggests that while high school students are as likely as adults to transmit the virus to others, children aged 6-11 are much less likely to do so. The authors show that three separate introductions of virus into three different elementary schools resulted in no further transmission of virus to other students or staff; consistent with results from Australia, Ireland, and a different French study.

However, these findings do not align with results from inquiries assessing viral load in children. After analyzing 3,303 COVID-19 patient sputum samples using two different PCR systems, German virologist Christian Drosten determined that children under the age of 19 produce virtually the same average levels of viral RNA as adults. Drosten argues that studies showing lower viral RNA levels in children relative to adults are subject to sampling bias because children are less likely to have symptoms and are therefore unlikely to be tested early in infection when virus levels in the nasopharynx are high. Instead, children are usually tested as contacts of index cases in symptom-triggered household studies, meaning that samples from children are obtained later in infection when virus levels in the nasopharynx are reduced. One way of validating this argument would be by analyzing children’s stool samples.    

Consistent with this suggestion, a recent report in Emerging Infectious Diseases suggests that both mildly symptomatic and asymptomatic children have high levels of viral RNA in the nose and saliva early during infection, but that these levels decline drastically within 1-2 weeks. In contrast, viral RNA levels in the feces remain high for more than three weeks after onset of symptoms. 

The data on transmission of SARS-CoV-2 from school-aged children to adults are limited because most schools have been closed since March. However, many child care centers in the US have remained open throughout the pandemic to care for the children of frontline workers and there have been no recorded coronavirus outbreaks directly linked to these facilities. Similar reports from Iceland, where extensive testing and contact tracing were implemented early during the pandemic, reveal only two documented cases of child-to-parent transmission, even though elementary schools and day care centers remained operational. 

School closures are a key intervention during epidemics of respiratory infections, and decisions surrounding the re-opening of schools this fall are fraught with uncertainty. While asymptomatically infected children are less likely to spread virus by coughing, and children have a smaller exhaled air volume than adults, they do engage in closer social contact with each other and are more physically active than adults. They are also more likely to put things in their mouths and share items. On the other hand, lengthy school closures have a negative impact on academic achievement and tend to increase educational inequalities because children of lesser means have less parental support and reduced access to resources for home learning. Only time will tell how school openings will affect the course of the pandemic. In the meantime, weekly testing of all students, followed by contact tracing and isolation of infected students and their contacts would be advisable.