Objectives: The authors sought to identify predictors of intensive care unit (ICU) admission among children hospitalized with bronchiolitis for ‡24 hours. Methods:The authors conducted a prospective cohort study during two consecutive bronchiolitis seasons, 2004 through 2006, in 30 U.S. emergency departments (EDs). All included patients were aged <2 years and had a final diagnosis of bronchiolitis. Regular floor versus ICU admissions were compared.Results: Of 1,456 enrolled patients, 533 (37%) were admitted to the regular floor and 50 (3%) to the ICU. Comparing floor and ICU admissions, multivariate ED predictors of ICU admission were age <2 months (26% vs. 53%; odds ratio [OR] = 4.1; 95% confidence interval [CI] = 2.1 to 8.3), an ED visit the past week (25% vs. 40%; OR = 2.2; 95% CI = 1.1 to 4.4), moderate ⁄ severe retractions (31% vs. 48%; OR = 2.6; 95% CI = 1.3 to 5.2), and inadequate oral intake (31% vs. 53%; OR = 3.3; 95% CI = 1.6 to 7.1). Unlike previous studies, no association with male gender, socioeconomic factors, insurance status, breast-feeding, or parental asthma was found with ICU admission. Conclusions:In this prospective multicenter ED-based study of children admitted for bronchiolitis, four independent predictors of ICU admission were identified. The authors did not confirm many putative risk factors, but cannot rule out modest associations.ACADEMIC EMERGENCY MEDICINE 2008; 15:887-894 ª