Conference Coverage

Distinguishing COVID-19 from flu in kids remains challenging


For children with COVID-19, rates of hospitalization, ICU admission, and ventilator use were similar to those of children with influenza, but rates differed in other respects, according to results of a study published online Sept. 11 in JAMA Network Open.

As winter approaches, distinguishing patients with COVID-19 from those with influenza will become a problem. To assist with that, Xiaoyan Song, PhD, director of the office of infection control and epidemiology at Children’s National Hospital in Washington, D.C., and colleagues investigated commonalities and differences between the clinical symptoms of COVID-19 and influenza in children.

“Distinguishing COVID-19 from flu and other respiratory viral infections remains a challenge to clinicians. Although our study showed that patients with COVID-19 were more likely than patients with flu to report fever, gastrointestinal, and other clinical symptoms at the time of diagnosis, the two groups do have many overlapping clinical symptoms,” Dr. Song said. “Until future data show us otherwise, clinicians need to prepare for managing coinfections of COVID-19 with flu and/or other respiratory viral infections in the upcoming flu season.”

The retrospective cohort study included 315 children diagnosed with laboratory-confirmed COVID-19 between March 25 and May 15, 2020, and 1,402 children diagnosed with laboratory-confirmed seasonal influenza A or influenza B between Oct. 1, 2019, and June 6, 2020, at Children’s National Hospital. The investigation excluded asymptomatic patients who tested positive for COVID-19.

Patients with COVID-19 and patients with influenza were similar with respect to rates of hospitalization (17% vs. 21%; odds ratio, 0.8; 95% confidence interval, 0.6-1.1; P = .15), admission to the ICU (6% vs. 7%; OR, 0.8; 95% CI, 0.5-1.3; P = .42), and use of mechanical ventilation (3% vs. 2%; OR, 1.5; 95% CI, 0.9-2.6; P =.17).

The difference in the duration of ventilation for the two groups was not statistically significant. None of the patients who had COVID-19 or influenza B died, but two patients with influenza A did.

No patients had coinfections, which the researchers attribute to the mid-March shutdown of many schools, which they believe limited the spread of seasonal influenza.

Patients who were hospitalized with COVID-19 were older (median age, 9.7 years; range, 0.06-23.2 years) than those hospitalized with either type of influenza (median age, 4.2 years; range, 0.04-23.1). Patients older than 15 years made up 37% of patients with COVID-19 but only 6% of those with influenza.

Among patients hospitalized with COVID-19, 65% had at least one underlying medical condition, compared with 42% of those hospitalized for either type of influenza (OR, 2.6; 95% CI, 1.4-4.7; P = .002).

The most common underlying condition was neurologic problems from global developmental delay or seizures, identified in 11 patients (20%) hospitalized with COVID-19 and in 24 patients (8%) hospitalized with influenza (OR, 2.8; 95% CI, 1.3-6.2; P = .002). There was no significant difference between the two groups with respect to a history of asthma, cardiac disease, hematologic disease, and cancer.

For both groups, fever and cough were the most frequently reported symptoms at the time of diagnosis. However, more patients hospitalized with COVID-19 reported fever (76% vs. 55%; OR, 2.6; 95% CI, 1.4-5.1; P = 01), diarrhea or vomiting (26% vs. 12%; OR, 2.5; 95% CI, 1.2-5.0; P = .01), headache (11% vs. 3%; OR, 3.9; 95% CI, 1.3-11.5; P = .01), myalgia (22% vs. 7%; OR, 3.9; 95% CI, 1.8-8.5; P = .001), or chest pain (11% vs. 3%; OR, 3.9; 95% CI, 1.3-11.5; P = .01).

The researchers found no statistically significant differences between the two groups in rates of cough, congestion, sore throat, or shortness of breath.

Comparison of the symptom spectrum between COVID-19 and flu differed with respect to influenza type. More patients with COVID-19 reported fever, cough, diarrhea and vomiting, and myalgia than patients hospitalized with influenza A. But rates of fever, cough, diarrhea or vomiting, headache, or chest pain didn’t differ significantly in patients with COVID-19 and those with influenza B.

Larry K. Kociolek, MD, medical director of infection prevention and control at Ann and Robert H. Lurie Children’s Hospital of Chicago, noted the lower age of patients with flu. “Differentiating the two infections, which is difficult if not impossible based on symptoms alone, may have prognostic implications, depending on the age of the child. Because this study was performed outside peak influenza season, when coinfections would be less likely to occur, we must be vigilant about the potential clinical implications of influenza and SARS-CoV-2 coinfection this fall and winter.”

Clinicians will still have to use a combination of symptoms, examinations, and testing to distinguish the two diseases, said Aimee Sznewajs, MD, medical director of the pediatric hospital medicine department at Children’s Minnesota, Minneapolis. “We will continue to test for influenza and COVID-19 prior to hospitalizations and make decisions about whether to hospitalize based on other clinical factors, such as dehydration, oxygen requirement, and vital sign changes.”

Dr. Sznewajs stressed the importance of maintaining public health strategies, including “ensuring all children get the flu vaccine, encouraging mask wearing and hand hygiene, adequate testing to determine which virus is present, and other mitigation measures if the prevalence of COVID-19 is increasing in the community.”

Dr. Song reiterated those points, noting that clinicians need to make the most of the options they have. “Clinicians already have many great tools on hand. It is extremely important to get the flu vaccine now, especially for kids with underlying medical conditions. Diagnostic tests are available for both COVID-19 and flu. Antiviral treatment for flu is available. Judicious use of these tools will protect the health of providers, kids, and well-being at large.”

The authors noted several limitations for the study, including its retrospective design, that the data came from a single center, and that different platforms were used to detect the viruses.

A version of this article originally appeared on

Next Article:

   Comments ()