SYNOPSIS: Two different studies published recently give a clear, consistent finding: About three-fourths of children hospitalized with SARS-CoV-2 do not have severe COVID-19-related illness but are merely identified as infected when subjected to screening tests. Surveys reporting the number or incidence of SARS-CoV-2-infected hospitalized children likely overestimate the actual burden of disease.

SOURCES: Webb NE, Osburn TS. Characteristics of hospitalized children positive for SARS-CoV-2: Experience of a large center. Hosp Pediatr 2021;11:e133-e141.

Kushner LE, Schroeder AR, Kim J, Mathew R. “For COVID” or “with COVID”: Classification of SARS-CoV-2 hospitalizations in children. Hosp Pediatr 2021;11:e151-e155.

The severity of illness with COVID-19 is classified in various ways, and hospitalized children often are automatically categorized as having severe disease. However, widespread screening for SARS-CoV-2 in hospitalized children makes it possible that children hospitalized for conditions unrelated to COVID-19 who are incidentally infected, with no symptoms attributable to the viral infection, might be reported as having “severe COVID-19 disease.” Two separate studies looking at the spectrum of disease severity of children hospitalized with SARS-CoV-2 infection were reported in the August 2021 issue of Hospital Pediatrics. Similarly designed and with similar findings, these reports help clarify the extent of illness caused by this infection.

First, Webb and Osburn studied SARS-CoV-2-positive children at a central California tertiary children’s hospital that has a referral population of 1.3 million children, 75% of whom have government-funded healthcare, and a majority of whom are identified as Hispanic. They retrospectively reviewed documentation about all patients younger than 22 years of age hospitalized with a positive SARS-CoV-2 antigen test (usually polymerase chain reaction) from May through September 2020; universal SARS-CoV-2 testing of all admitted patients was done during this time period. A total of 163 patients were identified and evaluated. Patients were assumed to have an incidental SARS-CoV-2 infection unrelated to the reason for hospitalization if they had no fever, no respiratory symptoms, and no gastrointestinal symptoms. Infections were categorized as “potentially symptomatic” if they were associated with fever or respiratory symptoms or gastrointestinal symptoms but without a requirement for respiratory support; this group included patients with diabetic ketoacidosis, appendicitis, and fever during the neonatal period with an admission to treat with antibiotics while ruling out serious bacterial infection. Patients were categorized as “significantly symptomatic” if they had respiratory or cardiac findings consistent with COVID-19 requiring respiratory support and/or intensive care.

Some other patients (17 of 163 overall infected patients in the study) were categorized by physician diagnosis (following Centers for Disease Control and Prevention [CDC] diagnostic criteria] as having multisystem inflammatory syndrome in children (MIS-C). Patients with MIS-C were excluded from subsequent analysis for the purposes of this paper, leaving 146 patients for evaluation. Overall, 58 (40%) of the 146 patients with acute SARS-CoV-2 infection were deemed to be “incidentally infected” (11 with a fracture, seven with seizures), 68 (47%) were deemed to be potentially symptomatic (with 25 of the 68 having appendicitis), and just 20 (14%) were deemed significantly symptomatic.

Significantly symptomatic patients were of statistically similar age (average 11 years) to those with incidental infection or potentially symptomatic infection (8 years). Approximately 90% of significantly symptomatic patients (and half of other patients) had medical comorbidities. There were four deaths among studied patients, with only one of the deaths being attributable to COVID-19. The “incidentally infected” and “potentially symptomatic” groups were similar for essentially all statistical analyses.

Second, Kushner and colleagues did a similar retrospective study, but from May 2020 to February 2021 (thus, including the winter respiratory season) and including only patients younger than 18 years of age (not 22 years, as in the other study) at a university-based quaternary children’s hospital in northern California.

They initially categorized patients as asymptomatic if they had no symptoms consistent with CDC descriptions of COVID-19, mild/moderate if they had symptoms attributable to COVID-19 but did not require supplemental oxygen, severe if they required extra oxygen but not pressure respiratory support, and critical if they required ventilation support or had sepsis or multi-organ failure. The investigators subsequently categorized patients as to whether COVID-19 was likely or unlikely to have prompted a need for hospitalization.

A total of 117 patients were included in the study cohort, 71% of whom identified as Latino, 16% of whom were immunocompromised, and 27% of whom required intensive care unit admission related to SARS-CoV-2. There were no deaths during the study period, but one included patient died of COVID complications shortly after data collection was completed.

For 55% of patients, COVID-19 was deemed to be the “likely” cause of hospitalization. Of the 117 total patients, 39% were “asymptomatic” (related to the SARS-CoV-2 infection), 28% had mild/moderate symptoms, 8% had severe illness, and 13% had critical illness; 12% had MIS-C.

Both research groups agreed that basing assessments of the extent by which COVID-19 is affecting children solely on the number of patients hospitalized with SARS-CoV-2 infection is inappropriate.

Although their classification systems were slightly different, they each showed similar rates of SARS-CoV-2-positive children being hospitalized for reasons totally separate from having COVID-19.


During a recent television interview, I was asked why scientists initially said that children were not adversely affected by COVID-19 and if the Delta variant was the reason children are being so severely affected now. The questions pointed out how misuse of data can lead to inappropriate conclusions. First, just because children are less often severely sick with COVID-19 than are adults, that did not mean that children were never adversely affected.

Second, related to the Delta variant or not, widespread screening reveals that many children are infected (often without symptoms), but it still is inappropriate to assign COVID-19 as the cause of hospitalization in all hospitalized children who happen to be infected.

Yes, children can be infected by SARS-CoV-2. Yes, many children infected by SARS-CoV-2 remain asymptomatic of their infection, even if they happen to be hospitalized for appendicitis or a fracture during the time they are asymptomatically infected. Yes, infected children can get sick when infected by SARS-CoV-2. Yes, children can die of COVID-19, even though at lower rates than seen in adults.

The data from these two new studies do make it clear that focusing on the rates at which hospitalized children are SARS-CoV-2-infected will overestimate the severity and effect of the pandemic. Many children identified by universal inpatient screening as infected by SARS-CoV-2 are not symptomatic with COVID-19 and are not hospitalized because of their coronavirus infection.

Whichever categorization scheme is used, 40% to 50% of children hospitalized with SARS-CoV-2 infection are not hospitalized because of or for that infection. Only 10% to 20% of pediatric patients hospitalized with SARS-CoV-2 infection are critically ill (either with the acute infection or with MIS-C). These data can help us better understand and explain the impact of the ongoing pandemic on children.

Another way to characterize the severity of COVID-19 in children would be to report on only the hospitalized children who seemed to clinicians to be hospitalized because of their SARS-CoV-2 infection and not merely the result of an incidental infection that was not causing symptoms.

Such was the case of another recent multicenter study.1 That retrospective study included 874 children (younger than 18 years of age) admitted from February 2020 to January 2021 to one of 51 collaborating hospitals with symptoms referable to COVID-19. The median length of stay was four days, with 46% requiring intensive care.

Overall, 1.4% did not survive the illness. Children requiring intensive care were older (10 years vs. 6 years), heavier (body mass index [BMI] 20.1 vs. 18.9), or had MIS-C (44% vs. 15%).

Asthma was a common comorbidity in children requiring intensive care for COVID-19.


  1. Webb NE, Osburn TS. Characteristics of hospitalized children positive for SARS-CoV-2: Experience of a large center. Hosp Pediatr 2021;11:e133-e141.