As countries start to emerge from shutdowns imposed during the coronavirus pandemic, the issue of how the virus affects children - and whether schools should be reopened - is taking on heightened importance.
Findings from relevant studies provide hopeful signs, though questions still remain. Here's what we know.
There are fewer cases of Covid-19, the disease caused by the coronavirus, among children compared to cases among adults, according to the US Centres for Disease Control and Prevention (CDC).
The agency says about 2 per cent of confirmed US cases of Covid-19 have been among people under the age of 18. That rate is 2.2 per cent in China, 1.2 per cent in Italy and 0.8 per cent in Spain.
But, as with the general population, epidemiologists say those rates likely do not include asymptomatic children, since people without symptoms are rarely tested for the infection.
Officials have also warned doctors to be on alert for cases of a rare life-threatening inflammatory syndrome associated with Covid-19 in children which affects blood vessels and organs.
It is reportedly similar to toxic shock syndrome and Kawasaki disease, which causes swelling and redness in blood vessels throughout the body, usually signalled by a lasting fever.
Dr Rosalind Eggo, an infectious disease modeller at the London School of Hygiene and Tropical Medicine (LSHTM), said she had seen some indications from research that children may not spread the coronavirus as much as adults.
"There are hints that children are less infectious but it is not certain," says Eggo, who sits on a panel that advises the British government about the transmission of Covid-19 among children and within schools.
For children who show no sign of being infected it was difficult to tell how contagious they were, Eggo said, though there was some evidence that there may be a lower possibility of infection from them.
"We need more studies to really pin this down as it is so important," she said.
[[nid:489088]]
Her research had shown that there was a much lower level of symptomatic infection in those under 20 years old - perhaps as little as 20 per cent of infections showing clinical symptoms.
John Edmunds, a member of Britain's Scientific Advisory Group for Emergencies (Sage), told the government that it was striking how children did not seem to play much of a role in spreading the novel coronavirus.
"For most respiratory viruses and bacteria they play a central role, but in this, they don't seem to," said Edmunds, a professor at LSHTM.
"There is only one documented outbreak associated with a school, which is amazing - you would normally expect most of the outbreaks to be associated with schools but yet in global literature there is only one documented study," Edmunds said, citing a study of a French secondary school. "It is pretty remarkable."
A recent study by the University of Queensland in Australia observed a collection of international family clusters of Covid-19 and found that children were the initial source of infection in less than 10 per cent of cases.
The report, produced and posted in April on an academic preprint platform, the SSRN, has been submitted to The Lancet medical journal but has not yet been peer-reviewed.
Several small studies in countries such as Iran and France have come to similar conclusions, as has the Netherlands' National Institute for Public Health and the Environment.
At least one study that looked at the amount of coronavirus in the body of Covid-19 patients, a measurement known as "viral load", found that the volume was not related to age.
[[nid:488974]]
Researchers at the Institute for Virology at Berlin's Charite University of Medicine in Germany said that an analysis earlier this year of nearly 4,000 positive test samples found that the very young did not differ significantly from adults in viral load, leading them to caution against unlimited reopening of schools.
But a separate analysis of that study carried out by the University of Zurich in Switzerland cautioned that the findings are difficult to interpret because of the statistical methods used and the small number of cases identified in children and adolescents.
"A reanalysis of summary data with a test for trend suggests that there is moderate, but not overwhelming, evidence for increasing viral load with increasing age," wrote lead author Dr Leonhard Held, professor of biostatistics at the university.
For the latest updates on the coronavirus, visit here.