Multistate surveillance for laboratory-confirmed, influenza-associated hospitalizations in children 2003-2004

Background: Increasing use of rapid influenza diagnostics facilitates laboratory confirmation of influenza infections. We describe laboratory-confirmed, influenza-associated hospitalizations in a population representing almost 6% of children in the United States. Methods: We conducted population-based surveillance for influenza-associated hospitalizations between October 1, 2003, and March 31, 2004, in 54 counties in 9 states (4.2 million children) participating in the Emerging Infections Program Network. Clinical characteristics, predictors of intensive care unit admission and geographic and age-specific incidence were evaluated. Results: Surveillance identified 1,308 case-patients; 80% were < 5 years and 27% were < 6 months of age. Half of the patients and 4 of 5 pediatric deaths did not have a medical indication for influenza vaccination and were outside the 6- to 23-month age group. Twenty-eight percent of case-patients had radiographic evidence of a pulmonary infiltrate, 11% were admitted to intensive care and 3% received mechanical ventilation. The median length of hospital stay was 2 days. Community-acquired invasive bacterial coinfections (1% of patients) were associated with intensive care admission (adjusted odds ratio, 16.9; 95% confidence interval, 5.0-56.8). Thirty-five percent of patients >= 6 months old had received at least one influenza vaccine dose that season. The overall incidence of influenza-associated hospitalizations was 36 per 100,000 children (range per state, 10 per 100,000 to 86 per 100,000). Conclusions: Influenza was an important cause of hospitalizations in children during 2003-2004. Hospitalizations were particularly common among children < 6 months of age, a group for whom influenza vaccine is not licensed. Continued surveillance for laboratory-confirmed influenza could inform prevention strategies.