BACKGROUND: Although rhinovirus (RV) respiratory infections trigger asthma exacerbations, the etiologic association between this virus and severe exacerbations, as well as the clinical characteristics of adults at risk for RV-associated asthma that necessitates hospitalization, have not been established. METHODS: During 1999-2003, we conducted a cohort study of 101 adults prospectively enrolled at hospital admission for an asthma exacerbation. Patient characteristics and frequencies of RV in nasal specimens were analyzed, by reverse-transcription polymerase chain reaction (RT-PCR), at asthma-related hospital admission and at a 3-month convalescent follow-up visit. RESULTS: RV was detected by RT-PCR in 21% of hospitalized patients over a 4-year period and in 1.3% of patients who returned for a 3-month follow-up visit. RV detection was strongly associated with hospitalization for asthma (adjusted odds ratio [OR], 15.1 [95% confidence interval {CI}, 1.88-121.4]). After adjustment for baseline asthma severity, RV-positive patients were more likely than RV-negative patients to be current smokers and nonusers of inhaled corticosteroids (ICSs) (adjusted OR, 11.18 [95% CI, 2.37-52.81]; P=.002). CONCLUSIONS: RV respiratory infection is an etiologic agent in severe asthma exacerbations necessitating hospitalization in adults. Compared with hospitalized patients with asthma who were RV negative, RV-positive patients were significantly more likely to be smokers and nonusers of ICSs.