To assess the possibility that clinical data available at the time of hospital admission for patients with respiratory syncytial virus infection could predict the length of hospitalization without antiviral therapy, the charts of all 102 patients admitted with laboratory documented respiratory syncytial virus infection during 1982 to 1985 were reviewed. Two thirds (65) of the patients remained hospitalized greater than three days. Prolonged hospitalization could have been predicted for 40 of the 65 patients (61.5%) based on the need for intubation and ventilation on admission (14 patients) or the presence of underlying cardiac or respiratory disease (28 patients). Of the infants hospitalized when they were less than 2 months of age, 40% required intubation and mechanical ventilation, although only 16% had underlying cardiac or respiratory disease. Mechanical ventilation was required at the time of admission or subsequently for 14% (5/36) of the 2- to 4-month-old infants with respiratory syncytial virus infection. Among the 56 patients without underlying cardiac or respiratory disease who did not require intubation at the time of admission, 25 (45%) remained hospitalized longer than three days (mean hospital stay for these 25 patients, regardless of age, was six days). Neither the respiratory rate nor the presence or absence of fever on admission was useful in predicting the likelihood of a longer hospital stay for the patients who did not require intubation. In addition, none of the initial laboratory data, including the WBC count, the percentage of band forms or polymorphonuclear leukocytes, the presence or absence of hypoxia, or the chest x-ray film findings could be correlated with a longer or shorter duration of hospitalization.(ABSTRACT TRUNCATED AT 250 WORDS)