In the public and scientific discourse, there seems to be establishing a consensus that -contrary to flu epidemics- the role of children and schools in the covid-19 epidemic is small. On what research does this consensus rely?

I found a reference to a study on the 2003 SARS outbreak (also a coronavirus) which couldn't find an impact of school closures on the evolution of that epidemic.

There are anecdotal stories about children who are infected and do not transmit the disease very easily.

Children do not feature heavily in the age-breakdown of people having tested positive for covid-19 or who tested positive for antibodies.

I can imagine researchers are mostly emboldened from that age-breakdown that I just mentioned, but what about the countering hypotheses that...

"many children who are positive for either covid-19 or its antibodies test negative because their bodies so easily overcome the disease (yielding both low virus concentrations throughout all stages of the disease and yielding low antibody concentration afterwards -> low concentrations are difficult to detect in a test)"

"Children may overcome the disease relatively quickly and unharmed, but in a society where the schools are not closed they interact so intensely with a large number of people that -despite their fast recovery- they still contribute a great deal to the overall R0 of that epidemic."

  • 2
    There is no real consensus. For example in the Netherlands as of 10 May, bars and restaurants are still to remain closed but primary schools are open again. In South Korea schools are still closed but bars and restaurants are open. To me this is the most surprising outcome, despite all the worldwide sharing of info every country draws different conclusions.
    – Ivana
    Commented May 14, 2020 at 15:25
  • 1
    @Ivana In Germany, primary schools have opened as well. They know that this presents an increased risk of a second wave, but they decided that closing schools further would do more harm than a possible resurgence. // Looking at policies is problematic because politicians have to consider multiple issues and not just the epidemiological aspect, while this question asks solely for epidemiological opinion.
    – Narusan
    Commented May 15, 2020 at 5:53

3 Answers 3


In the scientific discourse around Covid-19, I haven't seen statements that "the role of children and schools in the covid-19 epidemic is small". What I've seen is statements that "the role of children and schools alone in the Covid-19 epidemic is small", that is, closing schools alone will have little impact on the spread of the disease.

In order to keep person A from infecting person B, you need to break all possible transmission links between the two people. It's very rare for two people whose children attend the same school to have that as the only transmission route between them. Odds are, they'll also shop at the same stores, attend the same church, eat at the same restaurants, and so on, all of which also provide potential transmission routes.

Unfortunately, the word "alone" tends to get dropped when moving from scientific discourse to public discourse, giving the impression that re-opening schools is a low-risk activity. In actual practice, schools are a highly effective transmission route for most diseases, especially those with airborne or aerosol transmission.

  • The assumption here is clearly that children even if they become infected they won't shed the virus much, infecting other age groups, e.g. at home. That may or may not turn out that way... youtu.be/m6Hr69JH_wA?t=2329 Commented Apr 24, 2020 at 2:45
  • I'm sorry, but without references this comes across as too much ad lib and personal opinion. Your characterisation of how a total shutdown is 'required' to abate an epidemic sounds of to me: in a simple SIR model of epidemic I think that measures like closing schools independently have an effect on the infection parameter and the cumulative intensity of the epidemic depends continuously on that parameter, so closing schools while e.g. not closing barber shops should yield a slow-down in the epidemic.
    – 5th decile
    Commented Apr 24, 2020 at 17:13
  • @ThibautDemaerel, I've been trying to find the specific reference, but Covid-19 information is busy clogging every search engine out there right now, so it's hard to find. The context was an epidemiologist's response to a politician's suggestion that re-opening schools would only increase the death rate by 2%.
    – Mark
    Commented Apr 24, 2020 at 20:21

There is a recent (April 2020) article from the Lancet which reviews school closures data for both SARS and Covid-19. Partial summary is quoted here.

Modelling studies of SARS produced conflicting results. Recent modelling studies of COVID-19 predict that school closures alone would prevent only 2–4% of deaths, much less than other social distancing interventions.


This is what the AAP (pediatrics) also said about viral infection for young people.


In Pre-K, the relative impact of physical distancing among children is likely small based on current evidence and certainly difficult to implement. Therefore, Pre-K should focus on more effective risk mitigation strategies for this population.

However, the middle school case in Israel (and some elementary school infections), gives pause.

https://www.thedailybeast.com/israeli-data-show-school-openings-were-a-disaster-that-wiped-out-lockdown-gains https://www.haaretz.com/us-news/.premium-america-wants-to-reopen-schools-here-s-how-to-learn-from-israel-s-mistakes-1.9006697

In my blog, I mostly cover the science communication and planning side: http://ppeoptionsforschools.wordpress.com

But I do point to some references.

Exposure Bias

One of the biggest problems with the K-5 kids is that there may be selection bias, something I call exposure bias. Adults have many more contacts a day (say 30-100) and with random different people. K-5 may only contact their family. And, their family may try to limit their exposure to other people to protect them.

Google comes up with a lot of reports


The CDC says confirmed cases in kids is low:

In the United States, 2% of confirmed cases of COVID-19 were among persons aged <18 years.4 In China, 2.2% of confirmed cases of COVID-19 were among persons aged <19 years old.1 In Italy, 1.2% of COVID-19 cases were among children aged <18 years.2 In Spain, 0.8% of confirmed cases of COVID-19 were among persons aged < 18 years.5


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.