Abstract
A novel influenza virus emerged in the United States in spring 2009, rapidly becoming a global pandemic. Children were disproportionally affected by the novel influenza A(H1N1) pandemic virus [A(H1N1)pdm]. This retrospective electronic medical record review study aimed to identify clinical predictors of disease severity of influenza A(HIN1)pdm infection in paediatric patients. Disease severity was defined on an increasing three-level scale from non-hospitalized, hospitalized, and admitted to the intensive care unit (ICU). From April 2009 to June 2010, 696 children presented to Texas Children's Hospital's emergency department, 38% were hospitalized, and 17% were admitted to the ICU. Presenting symptoms associated with severe influenza were dyspnoea [odds ratio (OR) 5·82], tachycardia (OR 2·61) and fatigue (OR 1·96). Pre-existing health conditions associated with disease severity included seizure disorder (OR 4·71), obesity (OR 3·28), lung disease (OR 2·84), premature birth (OR 2·53), haematological disease (OR 2·22), and developmental delay (OR 2·20). According to model fitness tests, presenting symptoms were more likely to predict severe influenza than underlying medical conditions. However, both are important risk factors. Recognition of clinical characteristics associated with severe disease can be used for triaging case management of children during future influenza outbreaks.