AbstractBACKGROUND:Prior prediction models for length of stay (LOS) in bronchiolitis have focused more on birth‐ and disease‐related risk factors than on early hospital course factors, particularly common clinical markers including respiratory status and caloric intake.OBJECTIVES:1) Study the associations of various clinical markers and LOS; and 2) develop a LOS prediction model.DESIGN:Retrospective cohort study.SETTING:Children's Hospital of Wisconsin.PATIENTS:Inclusion criteria were: age <365 days old; admission between November 1, 2004 and April 15, 2005; final diagnosis of bronchiolitis; placement on the bronchiolitis treatment protocol; and lack of concurrent condition impacting LOS.RESULTS:During the study period, 272/347 infants admitted with bronchiolitis met inclusion criteria. On hospital day 2, infants in the prolonged LOS group (≥108 hours) had a significantly greater number of hours on supplemental oxygen, maximum supplemental oxygen use, minimum supplemental oxygen use, maximum respiratory rate, mean respiratory score, and number of times suctioned. They had significantly lower minimum oxygen saturation and caloric intake. Recursive partitioning demonstrated five variables (hours of supplemental oxygen, maximum respiratory rate, minimum supplemental oxygen use, gestation, and caloric intake) to predict short or prolonged LOS with an area under the receiver‐operator characteristic curve of 0.89/0.72 in the learning/test trees; sensitivity, 0.85; and specificity, 0.82.CONCLUSIONS:There are important differences between infants with bronchiolitis having short and prolonged hospital stays, including several clinical markers identifiable on hospital day 2. This model may be a useful prediction tool for targeting early interventions for high‐risk infants. Journal of Hospital Medicine 2011. © 2011 Society of Hospital Medicine