The research does not prove that infected children are contagious, but it should influence the debate about reopening schools, some experts said.

Coronavirus testing at a mobile clinic at the Walker Temple A.M.E. Church in south Los Angeles earlier this month.
Coronavirus testing at a mobile clinic at the Walker Temple A.M.E. Church in south Los Angeles earlier this month.Credit…Mario Tama/Getty Images

It has been a comforting refrain in the national conversation about reopening schools: Young children are mostly spared by the coronavirus and don’t seem to spread it to others, at least not very often.

But on Thursday, a study introduced an unwelcome wrinkle into this smooth narrative.

Infected children have at least as much of the coronavirus in their noses and throats as infected adults, according to the research. Indeed, children younger than age 5 may host up to 100 times as much of the virus in the upper respiratory tract as adults, the authors found.

That measurement does not necessarily prove children are passing the virus to others. Still, the findings should influence the debate over reopening schools, several experts said.

“The school situation is so complicated — there are many nuances beyond just the scientific one,” said Dr. Taylor Heald-Sargent, a pediatric infectious diseases expert at the Ann and Robert H. Lurie Children’s Hospital of Chicago, who led the study, published in JAMA Pediatrics.

The study is not without caveats: It was small, and did not specify the participants’ race or sex, or whether they had underlying conditions. The tests looked for viral RNA, genetic pieces of the coronavirus, rather than the live virus itself. (Its genetic material is RNA, not DNA.)

Still, experts were alarmed to learn that young children may carry significant amounts of the coronavirus.

“I’ve heard lots of people saying, ‘Well, kids aren’t susceptible, kids don’t get infected.’ And this clearly shows that’s not true,” said Stacey Schultz-Cherry, a virologist at St. Jude Children’s Research Hospital.

“I think this is an important, really important, first step in understanding the role that kids are playing in transmission.”

Jason Kindrachuk, a virologist at the University of Manitoba, said: “Now that we’re rolling into the end of July and looking at trying to open up schools the next month, this really needs to be considered.”

The standard diagnostic test amplifies the virus’s genetic material in cycles, with the signal growing brighter each round. The more virus present in the swab initially, the fewer cycles needed for a clear result.

Dr. Heald-Sargent, who has a research interest in coronaviruses, began noticing that children’s tests were coming back with low “cycle thresholds,” or C.T.s, suggesting that their samples were teeming with the virus.

Intrigued, she called the hospital lab on a Sunday and asked to look back at test results for the past several weeks. “It wasn’t even something we had set out to look for,” she said.

She and her colleagues analyzed samples collected with nasopharyngeal swabs between March 23 and April 27 at drive-through testing sites in Chicago and from people who came to the hospital for any reason, including symptoms of Covid-19.

They looked at swabs taken from 145 people: 46 children younger than age 5; 51 children aged 5 to 17; and 48 adults aged 18 to 65. To forestall criticisms that really ill children would be expected to have a lot of the virus, the team excluded children who needed oxygen support. Most of the children in the study reported only a fever or cough, Dr. Heald-Sargent said.

To compare the groups fairly, the team included only children and adults who had mild to moderate symptoms and for whom they had information about when symptoms began. Dr. Heald-Sargent left out people who didn’t have symptoms and who did not remember when they had started to feel ill, as well as those who had symptoms for more than a week before the testing.

The results confirmed Dr. Heald-Sargent’s hunch: Older children and adults had similar C.T.s, with a median of about 11 and ranging up to 17. But children younger than age 5 had significantly lower C.T.s of about 6.5. The upper limit of the range in these children was a C.T. of 12, however — still comparable to those of older children and adults.

“It definitely shows that kids do have levels of virus similar to and maybe even higher than adults,” Dr. Heald-Sargent said. “It wouldn’t be surprising if they were able to shed” the virus and spread it to others.

“One takeaway from this is that we can’t assume that just because kids aren’t getting sick, or very sick, that they don’t have the virus,” said one infectious disease expert.
“One takeaway from this is that we can’t assume that just because kids aren’t getting sick, or very sick, that they don’t have the virus,” said one infectious disease expert.Credit…Tom Brenner/Reuters

The results are consistent with those from a German study of 47 infected children between the ages 1 and 11, which showed that children who did not have symptoms had viral loads as high as adults’, or higher. And a recent study from France found that asymptomatic children had C.T. values similar to those of children with symptoms.

C.T. values are a reasonable proxy for the amount of coronavirus present, said Dr. Kindrachuk, who relied on this metric during the Ebola outbreaks in West Africa.

Still, he and others said that ideally researchers would grow infectious virus from samples, rather than test only for the virus’s RNA.

“I suspect that it probably will translate into meaning that there is more actual virus there as well, but we can’t say that without seeing the data,” said Juliet Morrison, a virologist at the University of California, Riverside

Some RNA viruses multiply quickly and are prone to genetic errors that render the virus incapable of infecting cells. Some RNA detected in children may represent these “defective” viruses: “We need to understand how much of that is actually infectious virus,” Dr. Schultz-Cherry said.

(The researchers said they did not have access to the type of high-security lab required to grow infectious coronavirus, but other teams have cultivated virus from children’s samples.)

The experts all emphasized that the findings at least indicate that children can be infected. Those who harbor a lot of virus may spread it to others in their households, or to teachers and other school staff members when schools reopen.

Many school districts are planning to protect students and staff members by implementing physical distancing, cloth face coverings and hand hygiene. But it’s unclear how well staff members and teachers can keep young children from getting too close to others, Dr. Kindrachuk said.

“Frankly, I just haven’t seen a lot of discussion about how that aspect is going to be controlled,” he said.

Observations from schools in several countries have suggested that, at least in places with mild outbreaks and preventive measures in place, children do not seem to spread the coronavirus to others efficiently.

Strong immune responses in children could limit both how much virus they can spread to others and for how long. The children’s overall health, underlying conditions such as obesity or diabetes, and sex may also influence the ability to transmit the virus.

Some experts have suggested that children may transmit less virus because of their smaller lung capacity, height or other physical aspects.

Dr. Morrison dismissed those suggestions. The virus is shed from the upper respiratory tract, not the lungs, she noted.

“We are going to be reopening day care and elementary schools,” she said. If these results hold up, “then yeah, I’d be worried.”