Nevertheless, a C.D.C. analysis of 2,572 cases found three deaths. Babies seem more vulnerable, but the data was incomplete.
Children make up a very small proportion of American coronavirus cases so far and are significantly less likely to become seriously ill than American adults, according to a preliminary report on the first wave of coronavirus cases in the United States. But some have become very sick, and at least three have died.
The study, published Monday by the Centers for Disease Control and Prevention, also reported that children appear less likely than adults to develop any of the major known coronavirus symptoms: fever, cough or shortness of breath. That could suggest that many children have mild or undetected cases of the disease and could be spreading the virus to others in their families and communities.
“We don’t think many kids get severe disease, so are kids transmitting the disease at a significant rate that’s going to propagate this outbreak?” asked Dr. Srinivas Murthy, an associate professor of pediatrics at the University of British Columbia, who was not involved in the study. The answer to that question, he said, will be important in deciding how to manage the pandemic, when to reopen schools and how to predict future waves of infection.
The C.D.C. study said that 2,572 of the nearly 150,000 confirmed coronavirus cases reported in the United States between Feb. 12 and April 2 were patients under 18. Their median age was 11. Those cases included 850 from New York City, the current epicenter of the outbreak in the United States, 584 from the rest of New York State and 393 from New Jersey.
There were some significant gaps in the data. The researchers said that important information — like whether the patients went to the hospital or what, if any, symptoms they had — was not available for many of the cases in the study. State and local health departments, many of which have been overwhelmed by the fast-moving pandemic, may be able to provide more data in the future, the report said.
Of the 745 cases with data on whether the child was hospitalized, 147 children — about a fifth — were reported to have been hospitalized. Among adults that rate is about a third, the report said.
Forty percent of pediatric hospitalizations, or 59 cases, were for babies under 1. Five of the 15 children admitted to intensive care were babies. Children with underlying health conditions also appeared to have a greater risk of hospitalization, the report said.
“Compared to other respiratory diseases, this is incredibly unique in the proportion of severely ill children,” Dr. Murthy said. “We would expect more hospitalization based on the number of kids that might get infected, and we’re not seeing that at all. And we still don’t know why.”
Although the C.D.C. report did not give details about the three deaths, Illinois has reported the death of an infant under 1 from the virus.
Only a third of the cases studied had information on whether the child was hospitalized; only 13 percent had information on whether the child had other underlying medical conditions; and fewer than 10 percent of the cases had information about the child’s symptoms.
Also, because many of the cases occurred very recently, shortly before the publication of the report, the researchers said their analysis “might underestimate severity of disease or symptoms that manifested later in the course of illness.”
Despite the gaps in the data, experts said the report, which is believed to be the largest study to date of confirmed cases of the coronavirus in children, could provide a useful early snapshot of general trends.
Scientists are exploring several theories about why children seem relatively protected from severe disease. A leading idea is that a receptor in human cells that the viral particles bind to, called the ACE2 receptor, is not expressed as prominently in young children or might be a different shape. If so, in children, the virus might have more difficulty attaching to and entering cells, a step that is necessary for the virus to replicate and spread throughout the body.
Another theory is that children’s immune systems do not respond as aggressively to the virus as adult immune systems do, a process that can spiral out of control and cause as much damage as the infection.
“I’m getting more convinced that there’s a physiological age-based reason rather than children are just healthy,” Dr. Murthy said.
The U.S. data echoes many of the findings in reports on pediatric cases in China, where the global outbreak began, but differences in China’s testing methodology make it difficult to compare results. One study of about 1,400 children under 16 in China found that fewer than half had coughs and only 42 percent had fevers, a much lower symptom rate than in adult patients.
In the American study, researchers had information about symptoms for 291 children. Of those, 73 percent had fever, cough or shortness of breath, compared with 93 percent of adult patients age 18 to 64.
The largest study to date of Chinese children, found that nearly 6 percent of the 2,143 confirmed or suspected cases of patients under 18 developed a severe or critical illness, and one 14-year-old boy with a confirmed diagnosis of coronavirus died. More than 60 percent of the 125 children who became severely or critically ill were age 5 or younger, the study reported. And 40 of those were infants less than 12 months old.
The U.S. study also found that 57 percent of the reported cases were in boys, a gender pattern that has been seen in some adult studies.