Skip Navigation

Influenza Associated Pediatric Deaths During 2003-2004 Flu Season

By Eric Toner, M.D., December 16, 2005

In an article published yesterday in the New England Journal of Medicine (N Engl J Med 2005;353:2559-67), researchers from the CDC influenza Special Investigations Team report on pediatric deaths from the 2003-2004 influenza season. The flu season that year was memorable for several reasons: flu struck earlier than normal, was more severe than it had been in the 3 years prior, and there was a relatively poor match between the vaccine and the circulating strain of flu virus (see:, accessed December 15, 2005). There were also many news reports of pediatric deaths. In response, the CDC gathered statistics on influenza-related deaths in children.

Among U.S. residents under 18 years of age with laboratory confirmed influenza, there were 153 deaths reported. The mortality rates were highest in children under age 6 months (0.88 per 100,000), but deaths occurred in children of all ages (overall mortality rate 0.21 per 100,000). Table 1, below, summarizes some of the most remarkable of the CDC’s findings:

Table 1: Summary of CDC statistics related to pediatric deaths from flu, 2003-2004

Number of children younger than 18 who died from flu in 2003-2004 season


Median age of children who died

3 years

Number of deaths within 3 days of symptom onset

45 (29%)

Number of deaths within 24 hours of symptom onset

8 (5%)

Number who died at home or in transit to the hospital

47 (31%)

Number who died in ED prior to hospital admission

16 (10%)

Number who died who had an identified underlying health risk condition

49 (33%)

Number who died who had no underlying health problems

70 (47%)

Number with underlying neurological and neuromuscular
disorders not previously recognized as high-risk conditions

49 (33%)

Median duration of illness

5 days

There were three broad causes of death: exacerbation of underlying chronic illness, invasive secondary bacterial infection, and fulminant illness following an initially mild illness. Table 2, below, presents a sample of the clinical diagnoses of those children who died. Of note, 24% of the children who died had a clinically significant bacterial co-infection. Approximately 50% of those infections were Staphylococcus aureus, and 50% of those infections were resistant to methicillin.

Table 2: Sample of clinical diagnoses in children who died from flu, 2003-2004



Laryngotracheobronchitis, tracheobronchitis or croup


Clinical sepsis or shock


Disseminated intravascular coagulation




Myocarditis or pericarditis


In his accompanying editorial, Dr. Raphael Dolin, from Harvard Medical School, points out that there are no approved antiviral drugs or vaccines for children younger than 6 months of age, the very group at highest risk of mortality from influenza. He suggests, however, that immunization of pregnant women may confer some immunity to newborns and infants. In addition, he points out the importance of vaccinating contacts and infants’ caregivers as a way to prevent infection in children.

Clinicians should be aware of the diverse presentations of influenza in children and the sometimes fulminant nature of the illness, even in those who are otherwise healthy.